a) b - École Polytechnique de Montréal

© Sepehr Ravati, 2010.
UNIVERSITÉ DE MONTRÉAL
NOVEL CONDUCTIVE POLYMER BLENDS
SEPEHR RAVATI
DÉPARTEMENT DE GÉNIE CHIMIQUE
ÉCOLE POLYTECHNIQUE DE MONTRÉAL
THÈSE PRÉSENTÉE EN VUE DE L’OBTENTION
DU DIPLÔME DE PHILOSOPHIAE DOCTOR (Ph.D.)
(GÉNIE CHIMIQUE)
AVRIL 2010

présentée par : RAVATI Sepehr
UNIVERSITÉ DE MONTRÉAL
ÉCOLE POLYTECHNIQUE DE MONTRÉAL
Cette thèse intitulée:
NOVEL CONDUCTIVE POLYMER BLENDS
en vue de l’obtention du diplôme de : Philosophiae Doctor
a été dûment acceptée par le jury d’examen constitué de :
M. FRADETTE Louis, Ph.D., président
M. FAVIS Basil, Ph.D., membre et directeur de recherche
Mme HEUZEY Marie-Claude, Ph.D., membre
M. SUNDARARAJ Uttandaraman (U.T.), Ph.D., membre

DEDICATED
To my parents
iii

ACKNOWLEDGEMENTS
I would like to express my gratitude to my advisor Professor Basil D. Favis for all our insightful
conversations. His thoughtful advice and enthusiasm for science have always inspired me. All
what I learned from him, not only academically but in ways to do research, is priceless.
There are many other people I need to thank who contributed to the success of this dissertation in
different ways. Among them, special thanks to François Normandin from IMI for his direct help
on assembling a particular setup to measure conductivity throughout the sample for my project.
Thanks to Carol and Jacques for technical help in providing the particular equipments for both
the layer-by-layer and blending process parts. Thanks also to Mélina for her assistance and
discussions on the rheological part. I would also like to thank Patricia Moraille for her assistance
in AFM, Éric Duchesne for introducing SEM to me and Jean-Philippe Masse for helping me with
FIB equipment and for helpful discussions.
My sincere thanks to my friends Sina Mortazy, Reza Shirazi, Nick Virgilio, Pierre Sarazin, and
Mehdi Maziane for their assistance and useful discussions in this thesis.
This thesis and all of my past academic success is due in large part to the support of my brothers:
Shahrooz, Babak, and Sasan, without whom I could not have this glorious moment in my life.
My deepest gratitude goes to my parents for their unconditional love and support. They taught
me the right way to live and no rewards without hard working by their own life.
iv

RÉSUMÉ
Cette thèse présente, pour la première fois, une étude complète des morphologies de contact
interfacial (engloutissement complet) pour des mélanges ternaires, quaternaires et quinaires de
polyaniline (PANI) avec quatre polymères commercialement disponibles : le polyéthylène à
haute densité (HDPE), le polystyrène (PS), le poly(méthyl méthacrylate) (PMMA) et le
polyfluorure de vinylidène (PVDF). La morphologie d’engloutissement complet est une structure
dictée par la tension interfaciale et est thermodynamiquement prévisible selon la théorie
d’étalement de Harkins. Ce type de microstructure est formé dans un mélange polymère ternaire
quand un des coefficients d’étalement, tel que calculé par les tensions interfaciales des paires de
polymères dans le système, a une valeur positive. L’extension aux systèmes quaternaires et
quinaires de définitions thermodynamiques similaires mène à la prédiction et à la formation de
mélanges à plusieurs composants ordonnés hiérarchiquement. Le contrôle de la composition des
composants de tels systèmes ordonnés hiérarchiquement, dans le cas des mélanges binaires,
ternaires, quaternaires et quinaires, permet de produire des structures à percolation multiple, dans
lesquelles toutes les phases sont pleinement interconnectées et interpénétrées. Cette approche est
utilisée pour atteindre l’objectif de ce travail de recherche qui est d’échafauder un ensemble de
techniques pour réduire le seuil de percolation des composants dans les mélanges. Le seuil de
percolation des phases est défini comme étant la formation de connectivité sur une longue
distance dans un système aléatoire. Les restrictions géométriques des phases dans des systèmes
auto-assemblés, par exemple dans des structures dictées par la tension interfaciale telles les
morphologies à percolation multiple, peuvent réduire le seuil de percolation des phases de
manière significative. Une autre partie de ce travail visait la préparation de substrats poreux à
très petite surface provenant de mélanges polymères à percolation multiple, suivie par la
déposition de PANI, un polymère conducteur, sur la surface interne poreuse par une technique
couche par couche.
Des mélanges polymères constitués de plusieurs composants, dont le HDPE, le PS, le PMMA, le
PVDF et le PANI, et, dans certains cas, le PP, le PEMA, et le PS-co-PMMA, ont été préparés
dans un mélangeur interne. Des échantillons en forme de disque ont été préparés dans une presse
v

chaude par moulage par compression dans le but d’en mesurer la conductivité. Certains
échantillons ont été de nouveau moulés par compression pour effectuer des tests de recuit
statique, dans le but d’augmenter la taille des pores. L’extraction par solvant a été utilisée pour
extraire divers composants et générer de la porosité dans le substrat. L’extraction par solvant
permet aussi de déterminer le pourcentage de continuité des différents composants du mélange.
Deux techniques qualitatives de caractérisation sont utilisées dans le but de mieux comprendre la
morphologie des mélanges polymères. La microscopie électronique par balayage (MEB) et la
microscopie par force atomique (MFA) accompagnée d’irradiation par faisceau ionique focalisé
(FIF) fournissent de l’information sur la morphologie des échantillons et sur l’épaisseur de la
structure multicouche construite. Après une préparation par microtomie cryogénique et une
déposition d’or-palladium, l’échantillon est gravé par FIF, révélant un contraste topographique
entre les phases. Une telle topographie permet de distinguer les phases et cette distinction est
quantifiée par MFA. Des mesures de conductivité ont été effectuées avec un électromètre à
travers l’épaisseur verticale des substrats.
Dans la première partie de cette étude, un dispositif conducteur solide à trois dimensions, avec
un faible seuil de percolation, est fabriqué en contrôlant les effets d’encapsulation multiple dans
un mélange polymère à cinq composants. Cette approche de structure à percolation multiple est
contrôlée thermodynamiquement selon la théorie d’étalement de Harkins, qui prévoit l’étalement
de toutes les phases, y compris la phase de polymère conducteur, à travers le mélange, un
mélange quinaire de PANI et de quatre autres composants (PE, PS, PMMA et PVDF)
sélectionnés pour présenter une structure à percolation multiple avec un seuil de percolation de
moins de 5 %. Afin de localiser le PANI au coeur de cette structure à percolation quadruple,
toutes les tensions interfaciales doivent satisfaire l’équation de Harkins et les phases doivent
avoir un comportement d’engloutissement complet avec la hiérarchie désirée
(HDPE|PS|PMMA|PVDF|PANI). Dans la première partie du travail, la morphologie et les
diagrammes de continuité détaillés des mélanges binaires, ternaires, quaternaires et quinaires de
PE, PS, PMMA, PVDF et PANI, sont étudiés systématiquement afin de déterminer les plages de
concentration qui permettront la formation de cette structure morphologique innovatrice,
l’encapsulation multiple. Il est démontré qu’à travers le contrôle de la composition des couches
intérieures et extérieures des structures en oignon des phases dispersées, la morphologie peut
vi

devenir une structure hiérarchique auto-assemblée à percolation multiple. Cette approche aide à
contrôler la morphologie des phases continues, menant à une restriction géométrique pour les
phases et réduisant abruptement le seuil de percolation des phases, alors que la conductivité du
mélange quinaire peut être augmentée de 10 -15 S cm -1 (PE pur) à 10 -5 S cm -1 à 5 % de PANI et
jusqu’à 10 -3 S cm -1 à 10 % de PANI. Il s’agit des plus hautes valeurs de conductivité jamais
rapportées à ces concentrations de PANI dans des mélanges préparés à l’état fondu.
Dans la deuxième partie de l’étude, les efforts ont été dirigés vers la création de très grands
substrats poreux dans le but de mieux comprendre la nature des morphologies à percolation
multiple. Ces substrats étaient destinés à servir de substrat tridimensionnel dans la technique de
dépôt couche par couche. Des mélanges polymères à percolation double et triple ont été préparés
à l’état fondu : HDPE/PS/PVDF, HDPE/PMMA/PVDF, et HDPE/PS/PMMA/PVDF. Il est
démontré que le recuit statique d’un échantillon à percolation multiple est un procédé rapide qui
mène à de très grandes tailles de pores. Puisque les structures à percolation multiple sont
constituées de phases continues assemblées dans un ordre précis, l’extraction d’une ou de
plusieurs phases permet la formation d’ultra-pores, à cause des restrictions géométriques. Un
recuit des échantillons suivi d’une extraction par solvant permet la création de substrats poreux
de HDPE entièrement interconnectés, avec une surface très faible. Suite à la préparation d’un
substrat idéal pour l’addition subséquente de polyaniline, des couches alternées de PSS et de
PANI sont déposées sur sa surface intérieure. Cette approche permet de réduire le seuil de
percolation à aussi peu que 0,19 % de PANI. De plus, selon la quantité de PANI déposée, la
conductivité du substrat poreux peut être contrôlée entre 10 -15 S cm -1 et 10 -3 S cm -1 . Le dépôt de
polyélectrolytes sur un substrat polymère poreux tridimensionnel a été étudié en détail et le profil
de dépôt massique montre un comportement oscillatoire suivant un motif en zig-zag. Les
expériences révèlent qu’un dépôt plus uniforme et des couches plus épaisses de PSS/PANI sont
produits en présence de sel dans la solution polyanionique. L’application d’une charge sur le
substrat peut constituer un autre paramètre de contrôle : une charge plus importante mène à des
valeurs de conductivité plus élevées (jusqu’à 10 -3 S cm -1 ). Dans la deuxième partie du travail,
l’emphase a été mise sur des substrats à très faible surface afin de déterminer le plus faible seuil
de percolation possible pour le polyaniline, mais des substrats à grande surface peuvent aussi être
préparés en utilisant cette approche.
vii

Dans la dernière partie de cette étude, des mélanges polymères ternaires montrent un
comportement de mouillage complet. Conceptuellement, il s’agit de l’état où un des composants
aura toujours tendance à séparer les deux autres complètement, et du point de vue
thermodynamique, deux des trois coefficients d’étalement binaires sont négatifs et l’autre est
positif, selon la théorie d’étalement de Harkins. Ce travail couvre l’étendue totale des états
morphologiques possibles pour un tel système préparé à l’état fondu, sur tout le diagramme de
composition ternaire. Un mélange polymère ternaire composé de HDPE, de PS et de PMMA est
choisi comme système modèle montrant un mouillage complet, alors que le coefficient
d’étalement positif λPS/PMMA = 2,6 mN/m indique que le PS sépare les phases de HDPE et de
PMMA. Quatre sous-catégories de morphologie peuvent être identifiées : a) matrice/phase
dispersée coeur-coquille; b) tricontinuité; c) bicontinuité/phase dispersée; d) matrice/deux phases
dispersées distinctes. La microscopie électronique, de même qu’une technique basée sur
l’irradiation par faisceau ionique focalisé combinée à la microscopie par force atomique, est
utilisée afin d’illustrer clairement et d’identifier les différentes phases. L’extraction par solvant et
la gravimétrie sont utilisées pour mesurer le degré de continuité des systèmes afin d’identifier
efficacement les régions de haute continuité. Les diagrammes de composition triangulaires sont
utilisés pour distinguer ces différentes régions morphologiques et les résultats sont interprétés à
la lumière des tensions interfaciales des différentes paires et de leurs coefficients d’étalement
subséquents. L’effet de la masse moléculaire et du ratio de viscosité sur la taille des phases des
différentes structures est également considéré. Il est démontré que typiquement, un plus haut
ratio de viscosité, de même qu’une plus haute tension interfaciale, mène à une plus grande taille
de phases.
viii

ABSTRACT
This thesis presents, for the first time, a comprehensive survey on the subject of self-assembled,
multi-encapsulated structures for ternary, quaternary, and quinary polymer blends demonstrating
complete wetting. The work targets the development of novel conductive devices possessing low
percolation threshold concentrations of conductive polymer. The blends studied are comprised of
conductive polyaniline and four commercial polymers: HDPE, PS, PMMA, and PVDF. A
complete wetting morphology is an interfacial tension driven structure which can be
thermodynamically predicted by the Harkins spreading theory. This type of microstructure in a
ternary polymer blend is formed when one of the spreading coefficients, as calculated from the
interfacial tensions of polymer pairs in the system, has a positive value. Conceptually, for a
ternary blend, complete wetting is the state where one of the components will always tend to
completely separate the other two. The development of similar thermodynamic definitions to
quaternary and quinary systems results in the prediction and construction of hierarchically
ordered multi-encapsulated, multi-component blends. Controlling the compositions of
components in such hierarchically ordered systems in ternary, quaternary, and quinary blends
allows for the preparation of multi-percolated structures in which all phases are fully
interconnected and interpenetrated. This approach is employed to attain the goal of this research
which is to build a foundation of methods to reduce the percolation threshold of components in
multi-component blends. The percolation threshold of phases is defined as the formation of longrange
connectivity in random systems. The geometrical restriction of phases in self-assembled
systems, for example in interfacial tension driven structures such as a multi-percolated
morphologies, can significantly reduce the percolation threshold of phases. Another part of the
work has focused on the preparation of ultra-low surface area porous substrates derived from
multi-percolated polymer blends, followed by the deposition of polyanilene conductive polymer
(PANI) on the internal porous surface using a layer-by-layer technique.
Polymer blends comprising various components including HDPE, PS, PMMA, PVDF, and
PANI, and in some cases PP, PEMA, and PS-co-PMMA are prepared in an internal mixer. Discshaped
samples for conductivity testing are compression molded in a hot press. Some of the
samples, after melt processing, are compression molded in a hot press to perform quiescent
ix

annealing to increase the phase sizes. Solvent extraction was used to extract various components
and generate porosity in the substrate. Selective solvent extraction also allows for the
determination of the percent continuity of the different components in the blend system. The
characterization, aimed at gaining insight on the morphology of polymer blends, is carried out
via two techniques. Scanning electron microscopy (SEM) and focused ion beam (FIB)-atomic
force microscopy (AFM) provide information about the morphology of the samples and the
thickness of the constructed multilayer. After cryogenically microtoming the sample and coating
it with a gold-palladium layer, the sample is etched by using FIB, demonstrating a topographic
contrast between phases. Such topography allows the phases to be distinguished and this is
quantified by AFM. Conductivity measurements were performed through the vertical thickness
of the substrates with an electrometer.
In the first part of this study, a solid, 3D, low percolation threshold conductive device prepared
through the control of multiple encapsulation effects in a five-component polymer blend system
through melt processing is fabricated. This multi-percolated structure approach is
thermodynamically controlled and is described by the Harkins spreading theory, demonstrating
spreading of all phases, including the conductive polymer phase, throughout the blend. A quinary
blend of PANI and four other components: polyethylene (PE), polystyrene (PS), poly(methyl
methacrylate) PMMA, and poly(vinylidene fluoride) (PVDF) are precisely selected to show
multi-percolated structure with a percolation threshold less than 5%. In order to locate PANI at
the core of this quadruple-percolated structure, all the interfacial tensions between phases must
satisfy the Harkins equation and show various complete wetting morphologies with the desirable
hierarchical ordering of the phases (HDPE|PS|PMMA|PVDF|PANI). The detailed morphology
and continuity diagrams of binary, ternary, quaternary, and finally, quinary systems for PE, PS,
PMMA, PVDF, and PANI, are progressively studied in order to systematically demonstrate the
concentration regimes resulting in the formation of these novel multiple-encapsulated
morphological structures. It is shown that through the control of the composition of the inner and
outer layers in the onion-type dispersed phase structures, the morphology can be transformed to a
hierarchical self-assembled, multi-percolated structure. This approach helps control the
morphology of continuous phases, leading to a geometrical restriction for the phases and sharply
reduces the percolation threshold of phases. In this fashion, the conductivity of the quinary blend
x

system can be increased from 10 -15 S cm -1 (pure PE) to 10 -5 S cm -1 at 5% PANI and up to 10 -3 S
cm -1 for 10% PANI. These are the highest conductivity values ever reported for these PANI
concentrations in melt processed systems.
In the second part of the study, a novel 3D porous polymeric conducting device is derived from
multi-percolated polymer blend systems. The work has focused on the preparation of ultra-low
surface area porous substrates followed by the deposition of polyanilene conductive polymer
(PANI) on the internal porous surface using a layer-by-layer self-assembly technique. The
approach reported here allows for the percolation threshold concentration of polyaniline
conductive polymer (PANI) to be reduced to values as low as 0.19%. Furthermore, depending on
the amount of PANI deposited, the conductivity of the porous substrate can be controlled from
10 -15 S cm -1 to 10 -3 S cm -1 . Ternary and quaternary multi-percolated systems comprised of high-
density polyethylene(HDPE), polystyrene(PS), poly(methyl methacrylate)(PMMA) and
poly(vinylidene fluoride)(PVDF) are prepared by melt mixing and subsequently annealed in
order to obtain large interconnected phases. Selective extraction of PS, PMMA and PVDF result
in a fully interconnected porous HDPE substrate of ultra low surface area and highly uniform
sized channels. This provides an ideal substrate for subsequent polyaniline(PANI) addition.
Using a layer-by-layer(LbL) approach, alternating poly(styrene sulfonate)(PSS)/PANI layers are
deposited on the internal surface of the 3-dimensional porous polymer substrate. The PANI and
sodium poly(styrene sulfonate)(NaPSS) both adopt an inter-diffused network conformation on
the surface. The sequential deposition of PSS and PANI has been studied in detail and the mass
deposition profile demonstrates oscillatory behavior following a zigzag-type pattern. The
presence of salt in the deposition solution results in a more uniform deposition and more thickly
deposited PSS/PANI layers. The conductivity of these samples was measured and the
conductivity can be controlled from 10 -15 S cm -1 to 10 -5 S cm -1 depending on the number of
deposited layers. Applying a load to the substrate can be used as an additional control parameter.
Higher loads result in higher conductivity values with values as high as 10 -3 S cm -1 obtained. The
work described above has focused on very low surface area porous substrates in order to
determine the lowest possible percolation threshold values of polyaniline, but high surface area
substrates can also be readily prepared using this approach.
xi

The final part of the work focuses on the complex morphological structures which can be
generated from a ternary polymer blend demonstrating complete wetting behavior. This work
examines the complete range of possible morphological states for such a system over the entire
ternary composition diagram as prepared by melt mixing. A ternary polymer blend comprised of
HDPE, PS, and PMMA is selected as a model system demonstrating complete wetting. The
positive spreading coefficient of λPS/PMMA=2.6 mN/m indicates that PS separates the HDPE and
PMMA phases. Four thermodynamically stable sub-categories of morphologies can be
identified: a) matrix/core-shell dispersed phase; b) tri-continuous; c) bi-continuous/dispersed
phase; and d) matrix/two separate dispersed phases. Electron microscopy as well as focused ion
beam irradiation and atomic force microscopy are used to clearly illustrate and identify the
various phases. Solvent extraction/gravimetry is used to examine the extent of continuity of the
systems so as to effectively identify regions of high continuity. Triangular compositional
diagrams are used to distinguish these various morphological regions and the results are
interpreted in light of the interfacial tension of the various binary combinations and their
subsequent spreading coefficients. The effect of the molecular weight and viscosity ratio on the
phase size of the various structures is also considered. It is shown that the viscosity of the phases
can influence the interfacial area between phases, but does not affect the morphological state
classification.
xii

CONDENSÉ EN FRANÇAIS
Dans ce travail, nous avons étudié la morphologie de mélanges à plusieurs composants préparés
à l’état fondu, suivi de gravure par solvant, afin de générer de nouveaux dispositifs conducteurs
avec un seuil de percolation extrêmement faible. Cette thèse mesure et élabore sur l’évolution de
la continuité dans les morphologies co-continues à percolation double, triple et quadruple, dans
le but de trouver le moyen de réduire considérablement le seuil de percolation de toutes les
phases dans le mélange.
L’utilisation de structures continues dans les mélanges polymères constitue une manière élégante
de réduire la continuité et/ou le seuil de percolation électrique d’un polymère conducteur. À cette
fin, le polyaniline (PANI) et quatre autres polymères, disponibles commercialement, avec les
tensions de surface, les polarités et les tensions interfaciales désirées, sont sélectionnés. Les
tensions interfaciales entre les différents composants se situent dans une plage variant entre 1
mN/m, pour le PVDF/PMMA, et 26,9 mN/m, pour le PANI/HDPE. La caractérisation
morphologique des échantillons est effectuée en utilisant la microscopie électronique à balayage
(MEB) et le faisceau ionique focalisé (FIF) couplé à la microscopie par force atomique (MFA).
Les mélanges polymères sont préparés à l’état fondu dans un mélangeur interne, à 50 tr/min
pendant 8 minutes à 200°C. L’extraction sélective des phases par solvant est utilisée, soit pour
aider à mieux distinguer les phases afin de recueillir de l’information qualitative, soit pour
produire des échantillons poreux. La combinaison du traitement par FIF et de la MFA permet de
distinguer clairement les phases à partir du contraste topographique résultant de la différence des
taux de gravure des matériaux par FIF. Afin d’étudier l’effet du ratio de viscosité dans les
mélanges ternaires, deux polyméthacrylates de méthyle (PMMA) sont choisis pour leurs masses
moléculaires différentes. Des matériaux poreux entièrement interconnectés sont utilisés comme
substrats pour la déposition alternée de PSS et de PANI.
Dans la première étape, trois composants, soit le polyéthylène à haute densité (HDPE), le
polystyrène (PS) et le PMMA sont soigneusement sélectionnés pour satisfaire les conditions
thermodynamiques. Le coefficient d’étalement positif λPS/PMMA = 2,6 mN/m, basé sur les trois
tensions interfaciales des paires de polymères (HDPE/PS (γPS/HDPE = 4,7 mN/m), PS/PMMA
(γPS/PMMA = 1,1 mN/m), et HDPE/PMMA (γHDPE/PMMA = 8,4 mN/m)), prévoit que la phase de PS
xiii

se localisera à l’interface des phases de HDPE et de PMMA dans une morphologie
d’engloutissement complet. Les résultats expérimentaux confirment la prédiction de la
morphologie par la théorie d’étalement, l’identification des phases étant clairement établie par la
topographie induite par la gravure par FIF et subséquemment quantifiée par l’analyse par MFA
en mode topographique. Un article publié, présentant pour la première fois une structure à
percolation double avec une couche de PS en forme de microgaine située à l’interface entre le
HDPE et le PMMA, est inclus en annexe. Il est démontré que par une combinaison de la théorie
classique de l’étalement et du contrôle de la concentration des phases, une couche uniforme de
PS est assemblée entre les phases avec un seuil de percolation extrêmement faible de 3 % de PS.
L’ajout d’une quatrième phase (polyfluorure de vinylidène, PVDF) au mélange ternaire avec un
engloutissement complet de HDPE/PS/PMMA, en autant que toutes les équations
thermodynamiques d’étalement soient satisfaites, mène à un ordre hiérarchique des phases,
HDPE|PS|PMMA|PVDF. Dans cette morphologie particulière, les phases de HDPE et de PVDF
deviennent les phases extérieures, alors que le PS et le PMMA se situent dans le milieu; le PS est
adjacent au HDPE, et le PMMA, adjacent au PVDF. Il est démontré que l’ordre des phases dans
tous les mélanges ternaires, HDPE|PS|PMMA, HDPE|PS|PVDF, HDPE|PMMA|PVDF, et
PS|PMMA|PVDF, est correctement prédit par l’équation de Harkins et est en accord avec l’ordre
des phases dans le mélange quaternaire HDPE|PS|PMMA|PVDF. La valeur positive du
coefficient d’étalement λPS/PVDF = 3,8 mN/m) dans le système ternaire HDPE/PS/PVDF indique
également que la phase de PS se situe entre le HDPE et le PVDF, ce qui est en accord avec
l’ordre du PS dans le mélange quaternaire. La théorie d’étalement de Harkins prédit que l’ajout
de PANI au mélange quaternaire HDPE|PS|PMMA|PVDF mènera à une structure ordonnée
hiérarchiquement HDPE|PS|PMMA|PVDF|PANI. Les coefficients d’étalement positifs des
différents mélanges ternaires montrent que tous les mélanges quaternaires, de même que le
mélange quinaire, de ces cinq polymères, devraient être ordonnés selon cette hiérarchie. Les
résultats expérimentaux sont en excellent accord avec ces prédictions.
Suite à la détermination théorique de l’ordre de toutes les phases, le contrôle de la composition
des phases permet de générer des morphologies innovatrices, telles la morphologie en « oignon »
et les structures à percolation multiple. Les micrographies obtenues par MEB confirment la
structure en oignon d’un mélange quaternaire HDPE/PS/PMMA/PVDF, avec une matrice de
HDPE et des gouttelettes de PVDF encapsulées par une première coquille de PMMA et une
xiv

deuxième coquille de PS. Les phases les plus importantes du mélange HDPE/PS/PMMA/PVDF
sont les première (HDPE) et dernière (PVDF) phases. La variation de la concentration de ces
deux phases, en maintenant les deux autres constantes, dicte l’évolution de la morphologie d’une
structure en oignon vers une structure à percolation multiple. Dans une structure à percolation
multiple, toutes les phases sont entièrement interconnectées et interpénétrées. Une région de
percolation multiple est une région de concentration dans laquelle toutes les phases dans le
mélange quaternaire HDPE/PS/PMMA/PVDF sont interconnectées, comme dans la région à
deux phases du diagramme de co-continuité d’un mélange binaire. Cette région est déterminée
par plusieurs paramètres, dont les ratios de viscosité et les tensions interfaciales entre les paires
de polymères. Afin de cerner quantitativement et précisément la région de percolation multiple,
l’extraction par solvant, suivie de mesures gravimétriques, a été utilisée pour obtenir le
pourcentage de continuité de chaque phase. La continuité maximale pour toutes les phases des
mélanges quaternaires a été obtenue à des compositions de 30/15/15/40 et 40/15/15/30
HDPE/PS/PMMA/PVDF. L’ajout de PANI à des mélanges quaternaires fortement continus
situés dans cette plage de composition mène à une morphologie à percolation quadruple avec les
phases ordonnées selon la hiérarchie HDPE/PS/PMMA/PVDF/PANI. Comme le PANI se situe à
la phase intérieure, il est porté à s’étaler entre les autres phases percolées, même à de faibles
concentrations. En conséquence, dans cette étude, pour la première fois, le seuil de percolation
de la phase intérieure (PANI) a été dramatiquement réduite à moins de 5 %. Le PANI utilisé dans
ce travail peut être mis en oeuvre à l’état fondu et contient 25 % massique de PANI pur et 75 %
massique de composé de zinc. Ceci indique que le volume réel de PANI pur présent dans le
mélange est inférieur à 5 % volumétrique. La morphologie du réseau de PANI est montrée après
l’extraction de toutes les phases, sauf le PANI, par le diméthylformamide, suivie d’un procédé de
lyophilisation, à partir d’un mélange 25/25/25/25 PS/PMMA/PVDF/PANI.
Dans une autre partie de ce premier article, l’effet de l’ajout de PS-co-PMMA au mélange
quinaire HDPE/PS/PMMA/PVDF/PANI à percolation quadruple est étudié. L’ajout d’un agent
compatibilisant pour deux phases adjacentes (ici, PS et PMMA) montre une réduction de la taille
des phases de PS et de PMMA dans le mélange. Puisque toutes les phases sont assemblées dans
un ordre hiérarchique, la réduction de la taille des phases de deux couches induit une réduction
de la taille des autres couches assemblées. L’ajout de 10% de PS-co-PMMA au mélange
30/30/30/10 PS/PMMA/PVDF/PANI, un système à encapsulation multiple, cause une réduction
xv

de toutes les tailles de phase dans le système encapsulé. Les micrographies des échantillons
contenant du copolymère montrent un réseau de PANI plus fin et plus uniforme, avec des
branches homogènes distribuées à travers l’échantillon.
Ce travail évalue l’effet du nombre de phases dans une structure à percolation multiple. Des
mélanges ternaires, quaternaires, quinaires et à six composants, contenant 5 % de PANI, sont
préparés. Les résultats indiquent que l’augmentation du nombre de composants réduit le seuil de
percolation de toutes les phases, incluant le PANI, à cause des restrictions géométriques des
phases. La valeur très faible de conductivité, près de 10 -12 S cm -1 , indique qu’aucun chemin
conducteur ne se forme dans le mélange ternaire PS/PMMA/PANI, puisque le 5 % de PANI est
présent sous forme de gouttelettes distribuées comme phase dispersée. Pour obtenir une phase de
PANI étalée, au moins quatre composants, dans une structure à percolation multiple, sont requis.
La conductivité augmente abruptement de presque six ordres de grandeurs avec l’ajout d’un
quatrième composant au système. L’ajout subséquent de phases n’augmente que légèrement la
conductivité, et la continuité du PANI, puisque quatre composants suffisent pour étaler la phase
de PANI à travers le mélange.
Dans le deuxième article, à travers le développement d’une morphologie à percolation multiple
accompagné d’une approche couche par couche (CPC), un nouveau dispositif conducteur
polymère poreux (DCPP) tridimensionnel est créé. Cette technique permet l’abaissement du seuil
de percolation du PANI conducteur à une valeur aussi faible que 0,19 %. Un substrat de HDPE,
poreux et entièrement interconnecté, avec une surface extrêmement faible, est préparé par la
création de systèmes à percolation double et triple, suivie d’un recuit et d’une extraction
sélective des phases par solvant. Tel que mentionné plus haut, le HDPE, le PS, le PMMA et le
PVDF produisent une structure à percolation multiple avec des phases ordonnées
hiérarchiquement, puisque les tensions interfaciales des différentes paires satisfont les conditions
positives des équations de Harkins. Des mélanges ternaires de 33% HDPE/33% PS/33% PVDF,
et 33% HDPE/33% PMMA/33% PVDF, et des mélanges quaternaires de 40% HDPE/30%
PS/10% PMMA/20% PVDF et 40% HDPE/10% PS/40% PMMA/10% PVDF ont été préparés.
Après un recuit statique de 15 minutes dans une presse chaude à une pression de contact très
faible, une augmentation substantielle dans la taille moyenne des phases est observée. Il est
démontré que la coalescence se produit beaucoup plus rapidement dans les systèmes à
percolation multiple que dans les mélanges binaires. L’extraction sélective par solvant dans les
xvi

structures à percolation multiple mène à la préparation de substrats avec un volume de vide plus
élevé. Ainsi, un substrat de HDPE poreux entièrement interconnecté à faible surface est préparé à
partir de morphologies à percolation double et triple. Les très grands pores des substrats de
HDPE favorisent aussi la pénétration de la solution PANI/PSS vers la zone poreuse
interconnectée. Jusqu’à 38 couches de PSS et de PANI ont été déposées sur la surface interne du
substrat polymère poreux tridimensionnel, en une conformation en réseau interdiffusé.
L’information sur la microstructure recueillie par MEB et par MFA dépeint une structure en
multicouche de polyélectrolyte relativement épaisse, jusqu’à 5,5 μm pour 38 couches de PSS et
de PANI sur la surface du substrat poreux de HDPE, dont la fraction de vide est de 66 %. Cette
épaisseur est attribuable à la diffusion du polyélectrolyte vers l’intérieur et vers l’extérieur dans
la structure multicouche au cours du procédé CPC. Le profil de dépôt massique du PANI/PSS en
fonction du nombre de couches montre une croissance non linéaire inhabituelle avec un
comportement oscillatoire. Ceci implique qu’après le trempage du substrat dans la solution de
PSS, la masse déposée augmente, et elle diminue ensuite après trempage dans la solution de
PANI, à cause de l’enlèvement partiel d’une couche précédemment adsorbée. Dans cet article, le
dépôt oscillatoire indique que les électrolytes PANI et PSS diffusent dans les couches déposées
auparavant. Les chaînes entrent en contact avec des chaînes semblables, et conséquemment, un
réseau interdiffusé de PANI et de PSS est généré. Puisqu’un sel est généralement ajouté dans les
procédés de dépôt CPC, l’effet de l’ajout de NaCl à 1 M à l’anion PSS dans la construction de la
structure multicouche est étudié. D’abord, le sel rend la structure multicouche plus uniforme.
Ensuite, le comportement oscillatoire est mis en évidence par un changement dans la
surcompensation des couches. Finalement, il est montré que des structures multicouches plus
épaisses sont construites en présence de sel dans le polyanion; cependant, la valeur de
conductivité est indépendante de la présence ou de l’absence de sel.
On en conclut que la surcompensation des charges à la surface de la structure multicouche est le
résultat de deux choses : l’ajout de sel à la solution, et la faiblesse du polycation (le PANI). De
plus, le déplacement de petits contre-ions et de segments de polymère chargés mène à des
polymères très gonflés, puisque des molécules d’eau liées aux ions y seront introduites.
Les mesures de conductivité démontrent que le seuil de percolation du PANI dans les dispositifs
poreux est atteint à un maximum de huit couches. Cette valeur correspond à 0,19 % massique de
PANI pour un substrat poreux fabriqué à partir d’un mélange de 33% HDPE/ 33% PS / 33%
xvii

xviii
PVDF, et à 0,28 % pour un substrat poreux fabriqué à partir d’un mélange de 33% HDPE/ 33%
PMMA / 33% PVDF. Il est montré que la conductivité de l’échantillon augmente avec
l’augmentation du nombre de couches de PSS/PANI déposées, jusqu’à un plateau de saturation,
qui est atteint après 32 couches déposées. Dans les matériaux poreux, la géométrie des pores
(incluant la taille et la distribution des pores) joue un rôle majeur dans la conductivité du
matériau, de même que la perfection du contact du matériau conducteur et la nature du matériau
qui remplit les pores. La compression des échantillons poreux peut être utilisée comme un autre
paramètre de contrôle, dans le but d’atteindre une large gamme de conductivités dans ces
dispositifs conducteurs poreux. L’application d’une compression sur les échantillons conducteurs
hautement poreux pousse et oblige les tiges et les murs du substrat à se déplacer vers l’intérieur.
Pour un substrat à 67 % de porosité, la conductivité augmente de 10 -8 S cm -1 à 0.002 S cm -1 après
le dépôt, mais pour un substrat à 60 % de vide, aucune déformation n’est observée.
Dans le dernier article, des mélanges ternaires à compositions variées de HDPE/PS/PMMA
représentent un cas de mouillage complet, avec la formation d’une couche
thermodynamiquement stable de PS entre les couches de HDPE et de PMMA. Quatre souscatégories
de morphologie thermodynamiquement stables existent pour le mélange
HDPE/PS/PMMA, selon la composition des phases : a) matrice/phase dispersée coeur-coquille
(60/10/30 HDPE/PS/PMMA); b) morphologie tri-continue (40/25/25); c) morphologie bicontinue/phase
dispersée (50/25/25); d) matrice/deux phases dispersées distinctes (20/60/20).
L’extraction par solvant, suivie de mesures gravimétriques, est utilisée pour étudier le degré de
continuité des composants. La morphologie matrice/coeur-coquille dans un mélange ternaire
A/B/C est constituée de gouttelettes de phase A encapsulée dans la phase B, le tout dans la
matrice C. Dans ce cas, l’augmentation de la concentration du coeur induit la coalescence des
phases de coeur, ce qui mène à la formation d’une morphologie tri-continue. Les structures tricontinues
peuvent être classées en deux catégories : type I et type II. Dans le premier cas, la
concentration de la phase centrale (PS) est assez faible pour qu’elle puisse se situer à l’interface,
mis à part quelques gouttelettes dans la phase extérieure (PMMA). Dans le cas du type II, avec
l’augmentation de la concentration de PS, davantage de gouttelettes de PS se situent dans le
PMMA, puisque l’interface HDPE/PS/PMMA est déjà saturée de PS. Au fur et à mesure que la
concentration de PS augmente, ces gouttelettes coalescent en des structures allongées, jusqu’à ce
que le PS forme aussi, au sein du PMMA, une phase pleinement continue et interconnectée. Un

autre sous-type de morphologie dans le cas du mouillage complet de HDPE/PS/PMMA est celui
d’une matrice avec deux phases dispersées distinctes. La morphologie bi-continue/phase
dispersée est constituée d’une structure co-continue pour deux phases, avec la troisième se
présentant sous forme de phase dispersée. Ce cas se produit quand la concentration de la phase
centrale, qui sépare les deux autres phases, augmente, ce qui donne une structure co-continue
pour la phase centrale avec une des deux autres phases. Un diagramme triangulaire de
composition et de morphologie, représentant diverses compositions de HDPE/PS/PMMA, est
préparé, afin de mieux comprendre la relation entre la composition et les différents états
morphologiques pour le mouillage complet, et les changements de phase (interconversion). Ce
diagramme montre les régimes de concentration pour : a) les mélanges tri-continus de type I et
de type II; b) une morphologie de type matrice/phases dispersées distinctes, avec le PS comme
matrice; c) une morphologie matrice bi-continue/phase dispersée, avec soit le HDPE, soit le
PMMA sous forme de gouttelettes; d) une morphologie matrice/coeur-coquille, avec soit le
PMMA ou le HDPE comme matrice. Ces régions sont déterminées par une combinaison de MEB
et de FIF-MFA, et grâce à l’extraction sélective et l’analyse gravimétrique, utilisées comme
technique quantitative pour repérer les points d’inversion. Des diagrammes triangulaires de
continuité sont présentés pour montrer la continuité des phases de PS et de PMMA. Dans un
mélange binaire, le seuil de percolation continue n’est qu’un point. Dans un mélange ternaire, il
devient une courbe, à cause du changement simultané des concentrations des deux autres phases
du mélange. Ce diagramme se compare aux courbes de co-continuité dans les mélanges binaires,
dans lesquelles la région de continuité double et le point d’inversion de phase sont distincts.
Deux courbes de composition, ou scénarios, qui recoupent de multiples états morphologiques à
travers le diagramme triangulaire à trois axes, sont choisis pour étudier l’effet de la concentration
des phases intérieure (HDPE) et centrale (PS) sur la morphologie de la phase centrale, et l’effet
de la composition de la phase intérieure (HDPE) et extérieure (PMMA) sur la morphologie de la
phase centrale (PS). L’observation la plus importante pour le scénario a), où la concentration de
PMMA est maintenue constante à 40 %, est qu’à cause des limitations thermodynamiques de
l’étalement, le système ne peut jamais former des gouttelettes de PS, même à de très faibles
concentrations de PS. Ainsi, même la plus petite quantité de PS dans le scénario a) présente une
structure en couches plutôt qu’en gouttelettes. Le scénario b) indique que, au pic du diagramme
xix

de continuité, une couche continue et parfaitement uniforme de 10 % de PS, avec une continuité
de 97 %, se localise à l’interface des phases co-continues de HDPE et de PMMA.
Il est démontré que des changements radicaux dans la viscosité de la phase PMMA (L-PMMA et
H-PMMA) dans un mélange PS/HDPE/PMMA de composition constante, ne modifie pas les
structures morphologiques; le PS sépare toujours le HDPE et le PMMA. La viscosité du PMMA
à haute masse moléculaire utilisé ici est presque 200 fois plus élevée que celle du PMMA à
faible masse moléculaire, ce qui correspond à une énorme augmentation du ratio de viscosités
PS/PMMA, mais ne change aucunement la classification morphologique du mélange. La phase
L-PMMA se présente sous forme de nombreuses gouttelettes encapsulées de PS, alors que le H-
PMMA se présente sous forme de quelques grosses gouttelettes encapsulées dans des coquilles
minces de PS. L’échelle des structures morphologiques a donc été significativement altérée par
la viscosité du PMMA, le système essayant de répondre à la réduction dramatique de l’aire
interfaciale des gouttelettes de H-PMMA. Il est démontré que la composition, la tension
interfaciale, et le ratio de viscosité jouent un rôle important dans la détermination de la taille des
phases dans les mélanges polymères ternaires. Le PMMA est beaucoup moins visqueux que le
PS, lui-même moins visqueux que le HDPE. Également, la tension interfaciale PS/PMMA est
beaucoup plus faible que celles des couples PS/HDPE et PMMA/HDPE. Pour le cas du mélange
50/40/10 HDPE/PS/PMMA, le PMMA est présent en faible concentration (10 %), le ratio de
viscosité PMMA/PS est faible, de même que la tension interfaciale PMMA/PS; ainsi, le PMMA
est distribué dans la phase de PS. Dans le mélange ternaire 50/25/25 HDPE/PS/PMMA,
l’augmentation de la concentration de PMMA induit la coalescence des gouttelettes de PMMA,
ce qui mène à la formation de grosses gouttelettes de PMMA, tel que montré par la topographie
obtenue par FIF-MFA pour ce mélange.
xx

TABLE OF CONTENTS
DEDICATION............................................................................................................................. III
ACKNOWLEDGEMENTS ....................................................................................................... IV
RÉSUMÉ ....................................................................................................................................... V
ABSTRACT ................................................................................................................................. IX
CONDENSÉ EN FRANÇAIS ................................................................................................. XIII
TABLE OF CONTENTS ........................................................................................................ XXI
LIST OF TABLES ............................................................................................................. XXVIII
LIST OF FIGURES .............................................................................................................. XXIX
NOMENCLATURE ................................................................................................................... XL
LIST OF ABBREVIATIONS .............................................................................................. XLIII
LIST OF APPENDICES ...................................................................................................... XLVI
CHAPTER 1 - INTRODUCTION AND OBJECTIVES ........................................................... 1
1.1 Introduction ..................................................................................................................... 1
1.2 Objectives ....................................................................................................................... 6
1.2.1 Specific objectives: ..................................................................................................... 7
CHAPTER 2 - LITERATURE REVIEW................................................................................... 9
2.1 Polymer Blends ............................................................................................................... 9
2.1.1 Miscible Polymer Blends ............................................................................................ 9
2.1.2 Immiscible Polymer Blends ...................................................................................... 12
2.1.2.1 Binary Polymer Blends ..................................................................................... 15
2.1.2.1.1 Co-continuous Morphology ........................................................................ 15
2.1.2.1.2 Interfacial Modification of Binary Blends .................................................. 20
xxi

2.1.2.2 Ternary Polymer Blends ................................................................................... 22
2.1.2.2.1 Composite Droplet Morphology ................................................................. 25
2.1.2.2.2 Double-Percolated Systems ......................................................................... 30
xxii
2.1.2.2.3 Effect of Composition on Ternary Blends with Complete Wetting
Morphology................................................................................................................... 33
2.1.2.2.4 Interfacially Modified Ternary Polymer Blends ......................................... 34
2.1.2.3 Quaternary Polymer Blend Systems ................................................................. 35
2.1.3 Porous Polymeric Substrates Prepared by Melt-Blending ........................................ 37
2.1.3.1 Controlling the Pore Size in the Substrate ........................................................ 37
2.1.3.1.1 Coarsening of Binary and Ternary Polymeric Systems .............................. 38
2.2 Percolation Theory ........................................................................................................ 40
2.2.1 Universality Systems ................................................................................................ 42
2.3 Conductive Materials .................................................................................................... 43
2.3.1 Electrically Conductive Polymers and Composites .................................................. 45
2.3.1.1 Conducting Polymer Composite Materials (CPCM) ........................................ 46
2.3.1.1.1 Parameters Influencing the Percolation Threshold of Conductive Fillers .. 49
2.3.1.1.2 Percolation Threshold of Carbon-Nanotube Polymer Composites ............. 54
2.3.1.1.3 Disadvantages of Conductive Fillers, Especially Carbon Black ................. 54
2.3.1.2 Intrinsically Conductive Polymers .................................................................... 56
2.3.1.2.1 Polyaniline ................................................................................................... 59
2.3.1.2.1.1 Binary PANI/Polymer or Ternary PANI/Polymer/Polymer Blends .... 59
2.4 Self-assembly Processes in Material Science ............................................................... 66
2.4.1 Layer-by-Layer Self-Assembly Technique ............................................................... 66
2.4.1.1 Fabrication of Conductive Multilayers by the LbL Method ............................. 70

xxiii
2.4.1.1.1 Charged Polymer Adsorption ...................................................................... 71
2.4.1.1.2 Polymer Adsorption on the Surface ............................................................ 73
2.4.1.1.3 Adsorption of Polyelectrolytes on the Surface of a Substrate ..................... 75
2.4.1.1.4 Various Mechanisms of Film Growth ......................................................... 78
2.4.1.1.5 Effect of Polyelectrolyte Concentration on Multilayers ............................. 79
2.4.1.1.6 Effect of Molecular Weight on Multilayers ................................................ 80
2.4.1.1.7 Effect of Salt on Multilayers ....................................................................... 81
2.4.1.1.8 Overcompensation of the Multilayer/Solution Interface ............................. 83
2.4.1.1.8.1 Overcompensation due to Difussion of Polyelectrolytes inside
Multilayers 83
2.4.1.1.8.2 Overcompensation due to the Addition of Salt .................................... 85
2.4.1.1.9 LbL Deposition on Curved Surfaces ........................................................... 88
2.4.1.1.10 Effect of Roughness on the Multilayer Thickness .................................... 88
2.4.1.1.11 Important Views on PANI/PSS LbL Adsorption ...................................... 89
2.4.1.1.12 Characterization Techniques for LbL Deposited Layers .......................... 90
CHAPTER 3 - ORGANIZATION OF THE ARTICLES ....................................................... 91
CHAPTER 4 - LOW PERCOLATION THRESHOLD CONDUCTIVE DEVICE
DERIVED FROM A FIVE-COMPONENT POLYMER BLEND ........................................ 95
4.1 Abstract ......................................................................................................................... 95
4.2 Introduction ................................................................................................................... 95
4.3 Experimental Methods ................................................................................................ 102
4.3.1 Materials ................................................................................................................. 102
4.3.2 Rheological Analysis .............................................................................................. 103
4.3.3 Sample Preparation ................................................................................................. 104
4.3.4 Solvent Extraction ................................................................................................... 105

xxiv
4.3.5 Characterization of Phase Morphology .................................................................. 106
4.3.5.1 Microtoming and Scanning Electron Microscopy (SEM) .............................. 106
4.3.5.2 Focused Ion Beam (FIB), Atomic Force Microscopy (AFM) ........................ 106
4.3.6 Conductivity Measurements ................................................................................... 107
4.4 Results and Discussions .............................................................................................. 107
4.4.1 Reducing the Percolation Threshold of PANI in a Multi-component Polymer
Blend 107
4.4.2 Self-Assembled Onion Morphology and the Hierarchical Ordering of Phases ...... 110
4.4.3 Generating Highly Continuous (Multiple-Percolated) Quaternary Polymer
Blends 115
4.4.4 Generating Highly Continuous Quinary Polymer Blends ...................................... 121
4.4.5 The Effect of Diblock Copolymer on the Morphology and Conductivity of
Multiple-Percolated Blend Systems .................................................................................... 125
4.4.6 The Effect of the Number of Components of Multi-Percolated Structures on
Conductivity ........................................................................................................................ 127
4.5 Conclusions ................................................................................................................. 131
4.6 Acknowledgements ..................................................................................................... 132
CHAPTER 5 - 3D POROUS POLYMERIC MATERIAL WITH LOW
PERCOLATION THRESHOLD OF CONDUCTIVE POLYMER PREPARED
USING LBL DEPOSITION ..................................................................................................... 136
5.1 Abstract ....................................................................................................................... 136
5.2 Introduction ................................................................................................................. 137
5.3 Experimental Methods ................................................................................................ 141
5.3.1 Materials ................................................................................................................. 141
5.3.2 Rheological Analysis .............................................................................................. 142
5.3.3 Substrate Preparation .............................................................................................. 143

5.3.4 Annealing Test ........................................................................................................ 144
5.3.5 Solvent Extraction ................................................................................................... 144
5.3.6 Layer-by-Layer Deposition ..................................................................................... 145
5.3.7 Filling the Porous Sample with a Resin .................................................................. 145
5.3.8 Characterization and Phase Morphology ................................................................ 145
5.3.8.1 Microtomy and Scanning Electron Microscopy ............................................. 145
5.3.8.2 Focused Ion Beam and Atomic Force Microscopy ......................................... 146
5.3.8.3 Conductivity Measurements ........................................................................... 146
5.4 Results and Discussion ............................................................................................... 146
5.4.1 Multiple-Percolated Morphology with Large Continuous Domain Phases ............ 146
5.4.2 Annealing the Samples in order to Increase Pore Size ........................................... 152
5.4.3 Continuity and Surface Areas of Substrates ........................................................... 154
5.4.4 Layer-by-Layer Deposition of PSS/PANI .............................................................. 156
5.4.5 Mass Deposition Dynamics of PSS/PANI .............................................................. 158
5.4.6 The Effect of Salt on the Mass Deposition Growth of PSS/PANI ......................... 162
5.4.7 Conductivity Measurement of the Conductive Porous Polymer Device ................ 165
5.4.8 Conductivity of the Conductive Porous Polymer Device under Loads .................. 167
5.5 Conclusions ................................................................................................................. 168
5.6 Acknowledgements ..................................................................................................... 169
5.7 References ................................................................................................................... 169
CHAPTER 6 - MORPHOLOGICAL STATES FOR A TERNARY POLYMER
BLEND DEMONSTRATING COMPLETE WETTING ..................................................... 172
6.1 Abstract ....................................................................................................................... 172
6.2 Introduction ................................................................................................................. 173
xxv

xxvi
6.3 Experimental ............................................................................................................... 177
6.3.1 Materials ................................................................................................................. 177
6.3.2 Rheological Analysis .............................................................................................. 178
6.3.3 Sample Preparation ................................................................................................. 179
6.3.4 Solvent Extraction ................................................................................................... 179
6.3.5 Characterization of Phase Morphology .................................................................. 180
6.3.5.1 Microtoming and Scanning Electron Microscopy (SEM) .............................. 180
6.3.5.2 Focused Ion Beam (FIB) and Atomic Force Microscopy (AFM) .................. 181
6.4 Results and Discussion ............................................................................................... 181
6.4.1 Matrix/Core-Shell Dispersed Phase Morphology ................................................... 182
6.4.2 Tri-continuous Morphology .................................................................................... 184
6.4.3 Matrix/ Two Separate Dispersed Phases Morphology ............................................ 185
6.4.4 Bi-Continuous/Dispersed Phase Morphology......................................................... 186
6.4.5 Triangular Composition/Morphology Diagram ...................................................... 187
6.4.6 Continuity, Co-continuity, and Tri-continuity of Phases ........................................ 191
6.4.7 Morphology Evolution due to Concentration Ratio of Two Components in
Complete Wetting Case ...................................................................................................... 194
6.4.8 Observations on the Effect of Viscosity on the Morphological State ..................... 199
6.4.9 Factors Influencing Phase Sizes.............................................................................. 201
6.5 Conclusions ................................................................................................................. 203
6.6 Acknowledgement ...................................................................................................... 204
6.7 References ................................................................................................................... 204
CHAPTER 7 - GENERAL DISCUSSIONS AND PERSPECTIVE..................................... 206
CONCLUSION AND RECOMMENDATIONS .................................................................... 209

xxvii
REFERENCES .......................................................................................................................... 211
APPENDICES ........................................................................................................................... 234

LIST OF TABLES
xxviii
Table 2-1. Relationship between concentration of the Polyanion and thickness of the
Monolayers in L-b-L deposition technique(Cheung, et al., 1997) .............................. 80
Table 4-1. Material Characteristics ............................................................................................. 103
Table 4-2. Surface Tension, Polar Contribution, Dispersive Contribution, Polarity, and Rate
of Surface Tension per Temperature for Homopolymers ......................................... 109
Table 4-3. Theoretical and Experimental Interfacial Tension for Polymer Pairs ....................... 109
Table 4-4. Spreading Coefficients in Different Ternary Blends and Predicted Order of
Phases in Blends ....................................................................................................... 112
Table 4-5. Number of Components and Compositions of Samples Shown in Figure 4-14 ........ 128
Table 5-1. Characteristic Properties of Homopolymers ............................................................. 141
Table 5-2. Composition of Polymer Components in Ternary and Quaternary Blends ............... 143
Table 5-3. Continuity of the Porous Morphology With and Without annealing (PS, PMMA
and PVDF extracted). ................................................................................................ 154
Table 5-4. Surface Area Measurements of Porous Polymer Substrates by BET ........................ 154
Table 5-5. Comparison of Surface Areas Reported in Previous Studies and in the Current
Work (last two rows). ............................................................................................... 156
Table 6-1. Material Characteristics ............................................................................................. 177
Table 6-2. Interfacial Tension of Polymer Pairs from (Reignier & Favis, 2000) and Positive
Spreading Coefficient of Ternary Blends ................................................................. 178

LIST OF FIGURES
xxix
Figure 1-1. a) Capillary vessels in eye(Malek et al., 2005) b) renal corpuscles tissue
(Rahmy, 2001) c)(Rahmy, 2001) Hypertrophied parietal cell d)(Rahmy, 2001)
Epithelial lining cells of a proximal convoluted .......................................................... 2
Figure 1-2. Human heart, longitudinal collagen fibers (bar = 50 µm)(Wang Gusukuma,
Prates, & Smith, 2004) ................................................................................................. 3
Figure 1-3. a), b), c) Some of the polymer blend samples observed by Veeco
lab(http://www.veeco.com/library/resources/Sample_Preparation/index.aspx),
d)three-phase contact morphology(Virgilio, Desjardins, L'Esperance, & Favis,
2009), e)two-phase contact morphology(Zhang, Ravati, Virgilio, & Favis,
2007), f) Spherulitic crystallization behavior of polymers(Feng & Kamal, 2005)
g) gold nanoparticles and their incorporation into polymer matrices(Meli &
Lennox, 2003) .............................................................................................................. 4
Figure 1-4. a), b) A model of the polyurethane foam made of interconnected rods, b) a
model of interconnected macroporosity, c)ultra-porous structure with 99%
porosity made of PDADMAC/BSA(Roy, Sarazin, & Favis, 2006) ............................ 4
Figure 1-5. MIT and Cambridge University scientists developed this tissue scaffold that
could help repair knees and other joints. The top section, indicated by the green
arrow, stimulates bone growth, while the lower half, marked by the orange
arrow, stimulates cartilage growth(http://web.mit.edu/newsoffice/2009/cartilage-
0511.html) .................................................................................................................... 5
Figure 2-1. Electrical conductivity versus PANI content of polymer/PANI binary
blends(Zilberman, et al., 2000d). ............................................................................... 20
Figure 2-2. Action of the diblock copolymer to reduce the pore size for 50/50 PLLA/PS
blends at copolymer concentrations of (a) 2.5%, (b) 5%, (c) 8%, (d) 12%, (e)
15%, and (f) 20%. Copolymer concentration shown is based on the weight of
polystyrene. The white bar indicates 1 micron(Sarazin & Favis, 2003). ................... 21

Figure 2-3. a), b), c) Ternary blends with partial wetting case, and d), e), f) ternary blends
with complete wetting case ........................................................................................ 23
Figure 2-4. SEM micrographs of the 80 (HDPE) /20 (PS/PMMA) I) a) fracture surface
etched with acetic acid for 2 min, 14%PS/86%H-PMMA; b) microtomed surface
xxx
etched 24 h with cyclohexane, 9%PS/91%H-PMMA; and c) microtomed surface
etched 24 h with cyclohexane, 9%PS/91%L-PMMA, II) Evolution of the shell
formation process with increasing PS content (vol. % based on the dispersed
phase) for the 80(HDPE)/20(PS/L-PMMA)(cryofractured samples after being
etched for 10 s with acetic acid). ................................................................................ 26
Figure 2-5. SEM micrograph of 70PA6/15PP/15PS blends: (a) cryofractured and (b)
chloroform extracted surfaces.(Harrats, et al., 2005) ................................................. 27
Figure 2-6. Dependence of the composite dispersed morphology on PS and PMMA
molecular weights. SEM photomicrographs of (a) HDPE/L-PS/L-PMMA;
PMMA is extracted by acetic acid and (b) HDPE/H-PS/H-PMMA; PS is
extracted by cyclohexane. The white bar denotes 1 μm(Reignier, et al., 2003). ....... 29
Figure 2-7. a, b) FIB-AFM images of HDPE/PS/PMMA ternary blends. The white line in
Figure 7b indicates the section analyzed below, and c) Continuity of PS or
PMMA as a function of the composition using the solvent dissolution
technique(Zhang, et al., 2007) ................................................................................... 30
Figure 2-8. Scanning electron micrograph of a) 40PA6/30PP/30PS; formic acid extracted
surfaces, and b) 40PA6/(25/5)(PP/PP–MA2)/(25/5)(PS/SMA2); formic acid
extracted surfaces (Harrats, et al., 2005) .................................................................... 31
Figure 2-9. Electrical conductivity versus LLDPE content for CoPA/LLDPE/PANI ternary
blends containing 10 and 20 wt% PANI(Zilberman, et al., 2000b) ........................... 32
Figure 2-10. Electrical conductivity versus LLDPE content for (PS/DOP)/LLDPE/PANI
blends containing 10 and 20 wt% PANI(Zilberman, et al., 2000b) ........................... 32
Figure 2-11. a) 70%HDPE/15%PS/15%PMMA, b) 15%PMMA/70%PS/15%HDPE(PMMA
extracted fracture surface), and c) 70%PMMA/15%PS/15%HDPE (PS extracted
fracture surface) (Guo, Packirisamy, et al., 1997) ..................................................... 33

Figure 2-12. SEM photomicrograph of a) 70%HDPE(A)/20%PP(B)/10%PS(C), b)
70%HDPE(A)/20%PP(B)/10%PS(C) + 0.5% S-E block copolymer, and c)
xxxi
70%HDPE(A)/20%PP(B)/10%PS(C) + 2% S-E block copolymer(Guo,
Packirisamy, et al., 1997) ........................................................................................... 35
Figure 2-13. SEM photomicrograph of a)
10%PMMA(D)/10%PS(C)/60%PP(A)/20%HDPE(B), b)
10%PMMA(D)/60%PS(A)/20%PP(B)/10%HDPE(C) , c)
60%PMMA(A)/20%PS(B)/10%PP(C)/10%HDPE(D), and d)
10%PMMA(D)/10%PS(C)/20%PP(B)/60%HDPE(A)(Guo, Gvozdic, et al.,
1997) .......................................................................................................................... 36
Figure 2-14. Coarsening process in co-continuous structure upon annealing.(a) the thin rod
merges/flows into thick ones at both sides, the thick rod merges into a much
thicker one at the same time; (b) the thin rod becomes thinner during the
merging process; (c) the thin rod breaks/splits; and (d) the residual parts
retract/shrink to the thick rods.(Yuan, 2005) ............................................................. 39
Figure 2-15. Experimental pore size growth (points) with the theoretical model (solid lines)
for different blends annealed at 200°C. (Yuan, 2005) ............................................... 40
Figure 2-16. Detail of a bond percolation on the square lattice in two dimensions with
percolation probability p=51(Stauffer & Aharony, 1994b) ....................................... 41
Figure 2-17. Logarithmic conductivity ladder locating some metals and conducting
polymers(Pratt, 2001)................................................................................................. 44
Figure 2-18. Simultaneous measurement of the nanocomposite (MWCNT/epoxy) electrical
conductivity as function of the nanotube weight fraction, performed in the liquid
state prior to curing at 0.1 s −1 and 20 °C. The rheological percolation threshold is
located at 0.1 wt% and the electrical one at 0.04 wt%(Du, et al., 2004). .................. 47
Figure 2-19. Approaching the percolation threshold in a binary blend ........................................ 49
Figure 2-20. Schematic illustrations of conductivity establishment for polymer composites
with (a) spherical fillers, (b) high aspect ratio fillers (fiber/wire) and (c)

percolation threshold (Vc) in a resistivity vs. filler loading plot(Wong, 2003-
xxxii
2008) .......................................................................................................................... 50
Figure 2-21. Transmission electron micrograph of silver nanowires(Wong, 2003-2008) ........... 51
Figure 2-22: Dependence of PE continuity fraction on the CB content in 5/95 and 10/90 .......... 53
Figure 2-23. TEM micrograph of a 50/50 PS/PMMA blend filled with 2 wt% CB. ................... 54
Figure 2-24. Plot of the total number of publications per polymer system versus the
minimum percolation threshold achieved with the respective system(Bauhofer &
Kovacs, 2009). ........................................................................................................... 55
Figure 2-25. Resistivity of carbon black/polymer composite as a function of carbon black
volume fraction(Rosner, 2000) .................................................................................. 55
Figure 2-26. The structure of a number of intrinsically conducting polymers (ICP). .................. 58
Figure 2-27. Transmission electron micrographs of extracted PANI-CSA/PMMA polyblend
films containing (a) Vol% = 0.5% , and (b) Vol% = 0.25% PANI-CSA(Yang, et
al., 1993a). .................................................................................................................. 62
Figure 2-28. a) Conductivity vs volume fraction(f) of PANI-CSA at T=300K and T=10K, b)
conductivity vs (f-fc) where percolation threshold (fc) is 0.3%. The solid lines
through the points correspond to t=1.99 at T=10K and t=1.33 at T=300K(Yang,
et al., 1993a). .............................................................................................................. 62
Figure 2-29: Conductivity vs. concentration of PANI/polymer prepared via a) dispersion
mixing, b) melt-blending(Narkis, et al., 2000a) ......................................................... 63
Figure 2-30. Electrical conductivity vs. CoPA content for (PS/DOP)/CoPA/PANI-pTSA
blends of various PANI-pTSA contents(Narkis, et al., 2000a) .................................. 64
Figure 2-31: Conductivity of PVDF/PANI and pure doped PANI as a function of PANI
content(Wt%) ............................................................................................................. 65
Figure 2-32: Conductivity of PANI/PVDF binary blend based on PANI concentration a)
(Malmonge, et al., 2006) and b) (Privalko, et al., 2005) ............................................ 65
Figure 2-33. LbL multilayers publications per year ..................................................................... 69

Figure 2-34. Schematic view of different polymer types: (a) linear homopolymers that are
the main subject of this review, (b) branched polymers, (c) charged polymers or
polyelectrolytes (PEs), and (d) a disordered copolymer with no specific order of
xxxiii
the different monomers. ............................................................................................. 71
Figure 2-35. Schematic view of the four scaling ranges in the Gaussian-persistent
regime(Netz & Andelman, 2003)............................................................................... 72
Figure 2-36. Different possibilities of substrates: (a) the prototype, a flat, homogeneous
substrate; (b) a corrugated, rough substrate. Note that experimentally, every
substrate exhibits a certain degree of roughness on some length scale; (c) a
spherical adsorption substrate, such as a colloidal particle. If the colloidal radius
is much larger than the polymer size, curvature effects (which means the
deviation from the planar geometry) can be neglected; (d) a flat but chemically
heterogeneous substrate; ............................................................................................ 74
Figure 2-37. The ‘tail’, ‘train’ and ‘loop’ sections of the adsorbing chain ................................... 75
Figure 2-38. Schematic drawing of single-chain adsorption. (a) In the limit of strong
coupling, and (b) in the case of weak coupling.......................................................... 76
Figure 2-39. Macromolecule with a) small persistence length, b) medium persistence length,
and c) large persistence length ................................................................................... 77
Figure 2-40. Schematic picture of a) the adsorbed polymer layer when the effective
persistence length is large, (b) Adsorbed layer for the case when the persistence
length is small. In this case the polymer forms a random coil with many loops ....... 77
Figure 2-41. Deposition of layer in LbL process a) in the absence of salt and b) in the
presence of salt ........................................................................................................... 81
Figure 2-42. Surface charge in LbL process as a function of salt concentration(Schlenoff &
Dubas, 2001) .............................................................................................................. 82
Figure 2-43. Penetration of polyelectrolyte chain inward and outward of a multilayer film ....... 84
Figure 2-44. Deplacement of polyelectrolyte ions with salt counter-ions .................................... 86

Figure 4-1. Schematic representation of (a) the complete wetting case where phase B
xxxiv
spreads between, and fully separates, phases A and C and (b) the partial wetting
case all phases are in contact with each other ............................................................ 97
Figure 4-2. Complex viscosity as a function of frequency at 190ºC for the homopolymers.
The complex viscosity of PANI is at 180ºC. ........................................................... 104
Figure 4-3. a) FIB-AFM image of the composite-droplet morphology of 30/10/60
HDPE/PS/PMMA showing PS as a layer separating HDPE and PMMA. Scan
size is 5μm×5μm. The bar to the right of the FIB/AFM micrograph indicates the
colours associated with the topographical height in nm. The white line in the
image indicates the section analyzed below, and b) SEM micrograph of a
45/10/45 HDPE/PVDF/PANI blend after extraction of the PVDF phase by
DMF. The voids show that PVDF was present as a layer separating HDPE and
PANI. ....................................................................................................................... 108
Figure 4-4. Schematic illustration of various encapsulated structures, a) dispersed phase B in
matrix A, b) core-shell morphology B/C in matrix A, c) quaternary onion
morphology B/C/D in matrix A, and d) quinary onion morphology B/C/D/E in
matrix A. .................................................................................................................. 111
Figure 4-5. SEM micrograph of a) matrix-droplet morphology of 80/20 PMMA/PS, b)
PVDF core-PMMA shell morphology in an HDPE matrix for 60/20/20
HDPE/PMMA/PVDF (PMMA extracted), and c) onion morphology in an
HDPE matrix for 60/13/13/13 HDPE/PS/PMMA/PVDF (PMMA extracted) ........ 113
Figure 4-6. SEM micrographs of 50/10/20/10/10 HDPE/PMMA/PS/PVDF/PANI after
extraction of (a) all phases except HDPE, (b) PS by cyclohexane (c) PMMA by
acetic acid, and (d) PS and PMMA by combining cyclohexane and acetic acid ..... 115
Figure 4-7. SEM micrographs and schematics of HDPE/PS/PMMA/PVDF after extraction
of PMMA by acetic acid for a) and b) onion morphology in an HDPE matrix for
60/13/13/13, c) and d) more elongated morphology of 50/17/17/17, and e) triplepercolated
morphology of 30/15/15/40 (microtomed surface). The voids show
that PMMA is present as a layer separating PS and PVDF. .................................... 117

Figure 4-8. SEM micrographs of various quaternary HDPE/PS/PMMA/PVDF blends after
xxxv
microtoming and extraction of the PMMA phase; (a) 10/15/15/60, (b)
20/15/15/50, (c) 30/15/15/40, (d) 40/15/15/30, (e) 50/15/15/20, (f) 60/15/15/10. .. 118
Figure 4-9. Continuity level of PS, PMMA, and PS+PMMA obtained by solvent
extraction/gravimetry for 6 blends comprised of HDPE/PS/PMMA/PVDF; (a)
10/15/15/60, (b) 20/15/15/50, (c) 30/15/15/40, (d) 40/15/15/30, (e) 50/15/15/20,
(f) 60/15/15/10. ........................................................................................................ 120
Figure 4-10. SEM micrographs of the multi-percolated 30/15/15/30/10 blend comprised of
HDPE/PS/PMMA/PVDF/PANI after extraction of (a) all phases except HDPE,
(b) PMMA by acetic acid (c) PS by cyclohexane, (d) PS and PMMA by
combining cyclohexane and acetic acid, and (e) PANI by chromic acid ................ 122
Figure 4-11. SEM micrographs of a), b) continuous PANI network of quaternary
25/25/25/25 PS/PMMA/PVDF/PANI blend after extraction of all phases by
DMF followed by a freeze drying process, and c, d, e) continuous PANI network
derived from a 15/20/15/25/25 PS/PS-co-PMMA/PMMA/PVDF/PANI blend
after extraction of all phases by DMF followed by freeze drying process. The
thickness of the PANI network walls are shown in Figure 4-11e. ........................... 123
Figure 4-12. a) Schematic representation of quinary multi-percolated morphology, the
yellow phase is PANI, the blue phase is HDPE, and all other phases are located
between them, b) schematic of a PANI network obtained after removal of all
phases except PANI, and c) a 3D image constructed from 2D SEM images of
the interconnected PANI network in the quinary 15/20/15/25/25 PS/PS-co-
PMMA/PMMA/PVDF/PANI blend after extraction of all other phases. ................ 124
Figure 4-13. SEM micrograph of a) 30/30/30/10 PMMA/PS/PVDF/PANI blend after
extraction of PS and PMMA, b) 20/20/20/30/10 PS/PS-co-
PMMA/PMMA/PVDF/PANI after extraction of PS and PMMA, c) 30/30/30/10
PMMA/PS/PVDF/PANI blend after extraction of PANI by chromic acid, and d)
20/20/20/30/10 PS/PS-co-PMMA/PMMA/PVDF/PANI after extraction of PANI
by chromic acid ........................................................................................................ 126

Figure 4-14. Conductivity of multi-percolated blends of ternary, quaternary, quinary, as well
as interfacially modified quaternary and quinary systems as a function of the
volume fraction of PANI. Note that point O is a co-continuous binary blend of
xxxvi
PS and PANI. ........................................................................................................... 129
Figure 5-1. 3D schematic showing the approach used to prepare the porous polymeric
conducting device via LbL deposition. (a) solid multi-percolated structure
comprised of HDPE, PS, PMMA and PVDF after melt blending and annealing;
(b) porous polymeric conducting device after extraction of PS, PMMA and
PVDF, followed by LbL deposition of PSS and PANI. .......................................... 147
Figure 5-2. (a) FIB-AFM topographic surface image of 40/20/40 HDPE/PS/PMMA. The
topographical height vs distance is shown to the right. The white line in the
image indicates the trace line used in the height vs distance analysis. and (b)
Scanning electron micrograph of 40/10/50 HDPE/PS/PMMA after extraction of
the PS phase by cyclohexane ................................................................................... 148
Figure 5-3. Scanning electron micrographs of binary and ternary samples after extraction of
all phases (except polyethylene) by DMF for and (a) 33/33/33
HDPE/PMMA/PVDF, (b) 33/67 HDPE/PVDF, and (c) 50/50 HDPE/PVDF ......... 149
Figure 5-4. Scanning electron micrographs of quaternary polymer blends for various
compositions compounded via melt-blending (a) 50/16/16/16
HDPE/PS/PMMA/PVDF after extraction of PMMA by acetic acid, (b)
50/20/10/20 HDPE/PS/PMMA/PVDF after extraction of PMMA by acetic acid,
(b) 50/20/10/20 HDPE/PS/PMMA/PVDF after extraction of PS by
cyclohexane, and d) 40/10/40/10 HDPE/PS/PMMA/PVDF after extraction of all
phases except HDPE ................................................................................................ 151
Figure 5-5. Scanning electron microscope images of the morphologies of the porous
structures showing the influence of annealing. The top row is without annealing,
the bottom row is after annealing. All components have been extracted except
HDPE. a),e) Sample (A); b),f) Sample (B); c),g) Sample (C); and d),h) Sample
(D). ........................................................................................................................... 153

Figure 5-6. a),b) Scanning electron micrographs of the morphology of sample (B) after
extraction of PMMA, PVDF and deposition of 38 PSS/PANI layers; c),d) SEM
and FIB-AFM of sample(B) after extraction of PMMA, PVDF and deposition of
xxxvii
38 PSS/PANI layers. Pores are filled with epoxy resin. .......................................... 157
Figure 5-7. Mass increase (%) indicating cumulative deposited mass as a function of number
of deposited PSS and PANI layers for solutions containing salt and without salt.
a) sample(A); b) sample (B); c) sample(C); and d) sample (D). ............................. 160
Figure 5-8. Schematic of exponential growth of PSS/PANI multilayer thickness on the
internal surface of the substrate. .............................................................................. 162
Figure 5-9. Mass increase (%) indicating cumulative deposited mass as a function of number
of deposited PSS and PANI layers for sample (A). a) solution containing salt,
and b) solution without salt. ..................................................................................... 165
Figure 5-10. Conductivity as a function of the number of deposited layers for various
samples at a 5 lb load. The effect of a 10 lb load is shown for two samples as
indicated by the arrows. ........................................................................................... 165
Figure 6-1. (a) schematic representation of the evolution of morphology in a binary blend,
(b) matrix/dispersed morphology, and (c) co-continuous morphology ................... 173
Figure 6-2. Schematic representation of (a) complete wetting: the case that one phase
locates between two other phases, and (b) partial wetting: the case that all phases
are in contact with each other .................................................................................. 176
Figure 6-3. Complex viscosity as a function of frequency at 190ºC for homopolymers ............ 179
Figure 6-4. FIB-AFM image of complete wetting morphology for ternary 40/20/40
HDPE/PS/PMMA blend showing the layer of PS at the interface of HDPE and
PMMA. Scan size is 25μm×25μm. The bar to the right of the FIB/AFM
micrograph indicates the colours associated with the topographical height in nm.
The white line in the image indicates the section analyzed. .................................... 182
Figure 6-5. (a) Schematic representation of the matrix/core-shell dispersed phase
morphology, (b) SEM micrograph of 60/10/30 HDPE/PS/PMMA blend after
extraction of PS by cyclohexane, (c) schematic representation of tri-continuous

xxxviii
morphology (type I), (d) FIB-AFM image of 40/20/40 HDPE/PS/PMMA
showing tri-continuous morphology. Scan size is 25μm×25μm. The bar to the
right of the FIB/AFM micrograph indicates the colours associated with the
topographical height in nm, (e) schematic representation of tri-continuous(type
II), (f) SEM micrograph of 30/30/40 HDPE/PS/PMMA after extraction of PS
by cyclohexane, (g) schematic representation of matrix/two separate dispersed
phases morphology, and (h) SEM micrograph of 10/80/10 HDPE/PS/PMMA
microtomed. ............................................................................................................. 184
Figure 6-6. a) Schematic representations of bi-continuous/dispersed phase morphology with
small droplets of white phase, b) FIB-AFM image of 50/40/10
HDPE/PS/PMMA showing bi-continuous/dispersed phase morphology with
small droplets of PMMA. Scan size is 8μm×8μm, (c) FIB-AFM image of
50/25/25 HDPE/PS/PMMA showing bi-continuous/dispersed phase morphology
with droplets of PMMA. Scan size is 4μm×4μm, d) schematic representations of
bi-continuous-dispersed phase morphology with small droplets of black phase,
e) FIB-AFM image of 10/30/60 HDPE/PS/PMMA showing bicontinuous/dispersed
phase morphology with droplets of HDPE. Scan size is
11.7μm×11.7μm, and f) SEM micrograph of 10/30/60 HDPE/PS/PMMA after
extraction of PS. Small droplets of HDPE are distributed in PS ............................. 187
Figure 6-7: Evolution of the morphology in a ternary A/B/C blend demonstrating complete
wetting as a function of the concentration of phases B and C. Phase A
concentration is held constant. ................................................................................. 188
Figure 6-8. a) Triangular concentration diagram showing the composition of the various
ternary HDPE/PS/PMMA blends examined in this study, b) various
morphological states for ternary HDPE/PS/PMMA, and c) triangular
concentration diagram showing the various regions of in the morphological
states for ternary HDPE/PS/PMMA ........................................................................ 190
Figure 6-9. Continuity diagrams for ternary HDPE/PS/PMMA blends. a) quantitative
continuity level of PMMA (percolation threshold line is shown in black), b)
quantitative continuity level of PS (percolation threshold line is shown in black),

and c) triangular concentration diagram showing the various regions of in the
morphological states for ternary HDPE/PS/PMMA after correction with the PS
xxxix
and PMMA continuity data. ..................................................................................... 193
Figure 6-10. Triangular concentration diagram showing two pathways: Continuity Scheme
(a), and Continuity Scheme (b). Scheme (a) is shown in Figure 6-11 and
represents the case where the concentration of PMMA is kept constant at 40% of
total while the concentration ratio of the PS and HDPE phases change. Scheme
(b) is shown in Figure 6-12 where the concentration of PS is kept constant at
10% of total while the concentration ratio of PMMA and HDPE phases change.
These two pathways were chosen since they pass through a range of different
morphological states. ............................................................................................... 195
Figure 6-11. PS continuity and the corresponding morphology as a function of HDPE
concentration following Scheme (a) from Figure 6-10. The PMMA phase
concentration is held constant at 40% volume fraction in all samples. In the
schematics, green is the PMMA phase, red is the PS phase and blue is the HDPE
phase......................................................................................................................... 196
Figure 6-12. PS continuity and the corresponding morphology as a function of HDPE
concentration following Scheme (b) from Figure 6-10. The PS phase
concentration is held constant at 10% volume fraction in all samples. In the
schematics, green is the PMMA phase, red is the PS phase and blue is the HDPE
phase......................................................................................................................... 198
Figure 6-13. SEM micrograph of morphology of (a) 35/40/25 HDPE/PS/L-PMMA after
extraction of L-PMMA by acetic acid, (b) 35/40/25 HDPE/PS/L-PMMA after
extraction of PS by cyclohexane, (c) 35/40/25 HDPE/PS/H-
PMMA(cryofracture), (d) 35/40/25 HDPE/PS/H-PMMA after extraction of
PMMA by acetic acid, and (e) 35/40/25 HDPE/PS/H-PMMA after extraction of
PS by cyclohexane, .................................................................................................. 200
Figure 6-14. FIB-AFM topographic surface image of HDPE/PS/PMMA at various
compositions, (a) 50/40/10, (b) 50/25/25, and (c) 10/30/60. The white lines in
the images indicate the sections analyzed for each blend. ....................................... 202

A cross-sectional area
a order parameter
Aij interfacial area
F Helmholtz free energy
f shape factor of pores
G conductance of porous material
G Gibbs free energy
G storage modulus
G0 conductance of the pure material
Gmix Gibbs free energy of the mixture
H enthalpy
Hmix enthalpy of mixing
I electric current
k Boltzmann constant
l length
l length of the sample
Lel Gaussian segment length
mi mass of component (i)
Mi Number of segments of polymer i
Mw weight average molecular weight
N number of molecules
NOMENCLATURE
xl

n number of moles
p concentration of the conductive material
P probability
pc percolation threshold concentration of conductive material
R resistance
R0 initial radius of droplet
Rel Gaussian blob size
S entropy
Smix entropy of Mixing
t dimensionless index for percolation equation
T absolute temperature
tan δ the ratio of storage modulus to loss modulus
V electric voltage
Vi volume of component (i)
Vs volume of segments
Wa work of adhesion
X P polarity
Greek letters
d
γ i
dispersive contribution to the surface tension of component (i)
γ ij interfacial tension between phases (i) and (j)
p
γ i
polar distribution to the surface tension of component (i)
xli

γ& shear stress
η ( & γ ) i
viscosity of phase i at specific shear stress
η zero shear viscosity
0
λ wavelength of the distortion in Tomotika’s equation
λ viscosity ratio
λij spreading coefficient of phase (i) over phase (j)
ρ density of component (i)
i
ρ resistivity
σ conductivity of the porous material
σ conductivity of the pure material
0
φ m maximum packing volume fraction
φ c
percolation threshold
Θ porosity
ξ relative linear dimension of the contact in conductive network
φi volume fraction of phase i
χ Flory-Huggins interaction
Ω number of configuration of molecules\
xlii

AFM atomic force microscopy
LIST OF ABBREVIATIONS
BET Brunauer Emmett and Teller nitrogen adsorption technique
CB carbon black
CLSM confocal laser scanning microscopy
CoPA copolyamide
CPC conductive polymer composite
CPCM conducting polymer composite material
CSA camphor-10-sulfonic acid
DBSA dodecylbenzenesulfonic acid
DMA dimethylacetamide
DMF dimethylformamide
DNNDSA dinonylnaphthalenedisulfonic acid
DOP dioctyl phthalate
EB emeraldine base
FIB focused ion beam
HA hyaluronan
HDPE high density polyethylene
H-PMMA high molecular weigth poly(methyl methacrylate)
ICP intrinsically conducting polymer
LbL Layer-by-Layer technique
LED light emitting diode
LLDPE linear low density polyethylene
xliii

L-PMMA low molecular weigth poly(methyl methacrylate)
MWCNT multi-wall carbon nanotube
OWLS optical waveguide lightmode spectroscopy
PA polyacetylene
PA6 polyamide 6
PAA poly(acrilic acid)
PAH poly(allylamine hydrochloride)
PANI polyaniline
PCL poly(ε-caplolactone)
PDADMAC poly(diallyldimethylammonium chloride)
PDADMAC polydiallyl dimethyl ammonium chloride
PEM polyelectrolyte multilayer
PEMA poly(ethyl methacrylate)
PET polyethylene terephthalate
PFU polyfuran
PGA poly(L-glutamic acid)
PI polyimide
PLL poly(L-lysine)
PLLA poly(L-lactide)
PMMA poly(methyl methacrylate)
PP polypropylene
PPP poly p-phenylene
PPY polypyrrole
PS polystyrene
xliv

PS-co-PMMA copolymer of PS and PMMA
PSS poly(sodium 4-styrenesulfonate)
PT polythiophene
PU polyurethane
PVA polyvinyl alcohol
PVC polyvinyl chloride
PVDF Polyvinyl difluoride
QCM quartz crystal microbalance
SEB styrene-ethylene-butylene diblock copolymer
SEBS styrene-ethylene-butylene-styrene triblock copolymer
SEM scanning electron microscopy
SMA styrene-maleic anhydride copolymer
SPM streaming potential measurements
SPR surface plasmon resonance
TSA toluenesulfonic acid
xlv

LIST OF APPENDICES
A- 1 Ultra-Low Percolation Thresholds in Ternary Co-Continuous Polymer Blends ............. 234
A- 2 Estimation of Conductivity of Porous Conductive Devices by Semi-Empirical Models 246
A- 3 Mechanism of Aggregation of Weak Polyelectrolytes .................................................... 250
A- 4 Relationship between Conductivity and Morphology of Ternary Polymer Blends ........ 252
xlvi

1.1 Introduction
CHAPTER 1 - INTRODUCTION AND OBJECTIVES
The study of the development of morphology in immiscible polymer blends is a domain of great
interest due to the influence of morphology on the physico-mechanical properties of the
materials as well as the control of the microstructure. Although obtaining desired properties(Paul
& Newman, 1978) by blending two or several components is a key challenge in polymer blends,
it is also important because novel complex microstructures for special applications can be
prepared. Polymer melt-blending is even more interesting, since by controlling the determining
parameters in the liquid-state, desirable structures with different geometries can be obtained.
Therefore, providing the required thermodynamic and kinetic conditions for a multi-component
blend processed via melt-blending results in the preparation of the desired self-assembly (the
notion that physical or chemical driving forces produce the necessary material microstructure
without the need for any external intervention) materials. For example, a binary blend of 40/60
polymer A/polymer B can show both matrix-dispersed morphology and co-continuous structure
depending on the mentioned properties. This implies that these systems, due to governing
properties, self-assemble to matrix-dispersed morphology or co-continuous morphology. Ternary
polymer blends in the polymer blend world is a critical case. It has been shown that due to
thermodynamic properties, three polymers can either be in contact with one another and make
three-phase contact, or one polymer can separate the second and third ones. These two cases are
seen in all the systems where liquids are in contact with each other. The most important ones are
detected in cell structures. Some of the examples are represented in Figure 1-1. More
complicated cases can be found in cell structures as phases hierarchically assembled in order
(Figure 1-2).
A key challenge in polymer blends is how to provide the system requirements to obtain desired
morphologies. The structures shown in Figure 1-1 are very close to the thermodynamically
controlled self-assembled structures made by polymer blends. In this case, the goal is designing
novel materials and finally replacing them with real materials or applying them in hightechnology
applications. Some of the important structures produced are introduced in Figure 1-3.
1

Particular morphologies can be developed by employing block copolymers as compatibilizers in
polymer blends and as templates for ordered structures.
a)
c) d)
Figure 1-1. a) Capillary vessels in eye(Malek et al., 2005) b) renal corpuscles tissue (Rahmy,
2001) c)(Rahmy, 2001) Hypertrophied parietal cell d)(Rahmy, 2001) Epithelial lining cells of a
proximal convoluted
b)
2

Figure 1-2. Human heart, longitudinal collagen fibers (bar = 50 µm)(Gusukuma, Prates, &
Smith, 2004)
d) e)
3

f) g)
Figure 1-3. a), b), c) Some of the polymer blend samples observed by Veeco
lab(http://www.veeco.com/library/resources/Sample_Preparation/index.aspx), d)three-phase
contact morphology(Virgilio, Desjardins, L'Esperance, & Favis, 2009), e)two-phase contact
morphology(Zhang, Ravati, Virgilio, & Favis, 2007), f) Spherulitic crystallization behavior of
polymers(Feng & Kamal, 2005) g) gold nanoparticles and their incorporation into polymer
matrices(Meli & Lennox, 2003)
In more recent works, novel complex microstructures for special applications are prepared by a
combination of melt-blending and other techniques. For example, porous templates can be made
by polymer melt-blending followed by selective extraction of one or several phases (Figure 1-4).
a) b)
Figure 1-4. a), b) A model of the polyurethane foam made of interconnected rods, b) a model of
interconnected macroporosity, c)ultra-porous structure with 99% porosity made of
PDADMAC/BSA(Roy, Sarazin, & Favis, 2006)
c)
4

One of the applications for porous biomaterial polymer blends is bone substitutions. The
proposed polymeric scaffolds are able to stimulate the bone regeneration process and to induce
the mineralization of the surrounding tissue. Scaffolds with high mechanical properties and high
porosity can be prepared in order to produce scaffolds for load bearing applications (Figure 1-5).
Figure 1-5. MIT and Cambridge University scientists developed this tissue scaffold that could
help repair knees and other joints. The top section, indicated by the green arrow, stimulates bone
growth, while the lower half, marked by the orange arrow, stimulates cartilage
growth(http://web.mit.edu/newsoffice/2009/cartilage-0511.html)
As well, hierarchical systems exhibiting a multi-level distribution of porosity can be widely
employed in tissue engineering applications. Polymeric scaffolds characterized by macro-pores
(100-500 μm) and micro-pores (10-50 μm) are useful as carriers for mesoporous (2-10 nm) silica
spheres (e.g. MCM-41), with the objective of loading them with drug molecules (ibuprofen).
These three pore sizes are very important in drug delivery systems. Whereas macro-pores can be
used for cell colonization and blood vessel access, micro-pores are necessary for the diffusion of
nutrients/fluids inside a graft, and meso-pores are exploited for drug uptake and release.
5

The scaffold could offer a potential new treatment for sports injuries and other cartilage damage.
Three dimensional (3D) modelling and analysis of the reconstruction of the specimens is
required to facilitate the study of the porous subjects and also to control and produce the
desirable structures.
1.2 Objectives
Devices made from semiconductor materials such as radios and computers are the foundation of
modern electronics. Semiconductor devices include the various types of transistors, solar cells,
and many kinds of diodes, which work with different conductivity values in a range from 10 -8 to
10 -2 S Cm -1 . Controlling the conductivity and resistivity of materials for various applications is a
challenging problem.
Polyaniline (PANI) is popular for its ease of preparation, good level of electrical conductivity,
and environmental stability. Green protonated emeraldine has a conductivity on a semiconductor
level in the order of 10 0 S cm –1 . Although the conducting form of PANI has a good chemical
stability combined with relatively high levels of electrical conductivity, PANI suffers from the
common disadvantage of conducting polymers, which is poor processibility. For this reason,
PANI is used as a filler in the preparation of conducting composites, and for the surface
modification of microparticles, powders, fibers, textiles, membranes, and porous substrates,
endowing them with new electrical, chemical, and surface properties. A proposed way to cope
with the poor processibility of conducting polymers is the preparation of PANI colloids, but this
technique is limited to special applications. Melt-processing of PANI was suggested by several
groups(Haba, Segal, Narkis, Titelman, & Siegmann, 2000; Narkis et al., 2000a; Segal, Haba,
Narkis, & Siegmann, 2001; Shacklette, Han, & Luly, 1993) to combine the desired properties of
conventional polymers and PANI. They were not successful in controlling the morphology of the
blend, and consequently were unable to reach a percolation threshold lower than 15%. Also, in
those cases the percolation threshold area is narrow and a resistive blend is quickly converted to
a conductive blend. Therefore, the conductivity value of the material is not controllable in the
range of semi-conductivity.
6

On the other hand, a novel tri-continuous structure derived from HDPE/PS/PMMA double-
percolated morphology was reported, with a percolation threshold of PS as low as 3%(Zhang, et
al., 2007). Though this model is not usable for PANI due to its high surface tension and polarity,
a similar way is sought to spread the PANI phase throughout the blend.
Since the conductivity of the blend depends on the volume fraction of the conductive phase and
its morphology, the conductivity value of a sample can be controlled by manipulating the
structure of PANI and its possible pathways in the blend. As discussed, it is very important to
obtain a material with the exact desired conductivity value for special applications. Also, it is
important to prepare samples in 3D bulk and porous structures. Most of the other preparation
techniques produce 2D samples.
The main objective of this work is developing novel and complex morphologies for the
preparation of highly controllable 3D conductive materials. The study will also analyse the
relationship between control of the multi-component blend morphology and reduction of the
percolation threshold of the conductive phase.
1.2.1 Specific objectives:
• Understand and develop the morphology of ternary, quaternary, and quinary polymer
blends with complete wetting structure;
• Study the development of multi-percolated (double, triple,…) morphologies to obtain the
lowest possible continuity and conductivity percolation threshold in ternary, quaternary, and
quinary polymer blends;
• Evaluate the potential of both melt-blending/solvent extraction/LbL and multi-processing
techniques to reduce the conductivity percolation threshold of a conductive polymer with high
surface tension and polarity;
• Study the effect of interfacial modifier on the continuity and conductivity of multipercolated
morphology in quaternary blends;
7

• Evaluate the thermodynamic and composition parameters on continuity diagrams of
binary, ternary, quaternary and quinary systems in multi-component blends with multiple-
encapsulated morphological structures;
• Develop a method based on the Harkins equations set to predict the complex
morphologies in multi-component blends;
• Understand the mechanism and effect of parameters on the self-assembly LbL technique
and the formation of deposited multilayers for a weak polycation and a strong polyanion.
8

2.1 Polymer Blends
CHAPTER 2 - LITERATURE REVIEW
One of the most important ways to make novel, high performance polymer materials over the last
40 years has been polymer blending, which has found applications in many different scientific
fields. By using this method, many properties such as electrical, mechanical, physical and other
properties can be combined in a material and adjusted to the needs of particular end-use
applications (Hara & Sauer, 1998), although blends are also used for manufacturing polymeric
materials at lower cost or with improved processing behaviour. Polymers are macromolecules
with high molecular weight and particular structures; thus, most of them do not have favorable
interactions with each other.
There are two different groups of polymer blends: miscible and immiscible polymer blends. The
former are homogeneous and stable with intermediate properties. The latter group represents the
majority of polymers, with morphology-dependent properties. When polymers are mixed, a
phase-separated mixture is formed.
2.1.1 Miscible Polymer Blends
Development of thermodynamics of binary mixtures of low molecular weight liquids followed
by study of miscibility of a binary polymer blend started in the 19 th century. For over 50 years,
the cornerstone of polymer solution thermodynamics has been the Flory-Huggins model.
Theoretical equations of state for polymer liquids were explained in the mid-1960’s by Flory and
co-workers (Flory, 1965; Flory, Orwoll, & Vrij, 1964).
Thermodynamics determine the equilibrium states that can be achieved for any given set of
conditions when two polymers A and B are mixed together. Gibbs(Gibbs, 1876) formulated the
stability of multiphase systems in terms of the quantity G defined as follows:
9

Equation 2-1.
G = H − TS
where H is enthalpy, T is absolute temperature, and S is entropy. The necessary condition for
two substances A and B to be miscible is that the Gibbs free energy of the mixture must be lower
than the sum of the Gibbs free energies of the separate constituents, or in other words,
Δ mix
G < 0 . Therefore, the necessary condition can be extended to:
Equation 2-2. G = ΔH
− TΔS
< 0
Δ mix
mix
mix
where ΔH is the enthalpy of mixing and mix
Δ S is the entropy of mixing. Enthalpy and entropy
mix
of mixing have been modeled extensively. The combinatorial entropy is the most important
factor leading to miscibility in low molecular weight materials. For higher molecular weight
components, the TΔ S term is small and the enthalpy of mixing can dominate the miscibility.
mix
However, the Equation 2-2 is not a sufficient requirement, as the following expression must also
be satisfied:
2 ⎛ ∂ ΔG
⎞
Equation 2-3. ⎜ m ⎟
⎜
> 0
2 ⎟
⎝ ∂φi
⎠
Negative value of this equation can yield a region of the phase diagram where the mixture
separates into a phase rich in component A and another phase rich in component B(Denbigh,
1971). The entropy of mixing for dissimilar components can be determined(Clausius, 1864) with
the Boltzmann relationship as:
T , P
Equation 2-4. Δ = k ln Ω
S m
k is the Boltzmann constant and Ω represents the number of configurations or the summation of
combinations of arranging N1 and N2 molecules into a regular lattice of N cells:
Equation 2-5.
Ω
=
( N + N )
1
N ! N
1
2
2!
!
10

Based on Sterling’s approximation:
Equation 2-6. ln N! = N ln N − N
Installation of Equations 2-5 and 2-6 in the Equation 2-4 yields:
Equation 2-7. Δ = −k
( N ln x + N ln x )
where
x
1
N1
= and
N
x
2
S m
1
1
2
2
N 2
= . For mixing of polymer components, calculation of entropy of
N
mixing can be developed to(Flory, 1942, 1944; Huggins, 1942):
Equation 2-8. ⎟ ⎛ φ1
φ2
⎞
ΔS = − ⎜
m kV ln x1
+ ln x2
⎝V1
V2
⎠
where V is volume of each polymer molecule and φ 1 represents the volume fraction of phases.
The enthalpy of mixing was expressed by Flory(Flory, 1953) as:
Equation
ΔH
m =
φ kTN
2
1
χ
12
where χ12 is the Flory-Huggins interaction parameter and characterizes the interaction energy per
solution molecule divided by kT. Consequently, the free energy of solution for low molecular
weight molecules was expressed by Meyer-Flory-Huggins as:
Equation 2-9. Δ = kT ( N ln φ + N ln φ + χN
φ )
G m
1
1
McMaster found that volume changes during the mixing of polymers should be considered in the
represented formulation(McMaster, 2002).
Scott extended the calculation of free energy to polymer blends by the following relationship:
Equation 2-10. ⎟ ⎛ V ⎞⎛
φ1
φ2
⎞
ΔG =
⎜
⎟
⎜
m
kT lnφ1 + lnφ2
+ χφ1φ
2
⎝Vs
⎠⎝
M1
M 2
⎠
2
2
1
2
11

where the number of segments of the two polymers are represented by M1 and M2, and Vs is the
volume of the segments(Scott, 1949).
2.1.2 Immiscible Polymer Blends
Most polymer blends are immiscible, showing phase-separated mixture. Morphology of
multiphase polymer blend materials can be controlled by controlling the interfacial chemistry
and the microstructure. Three parameters determine the properties of these multiphase materials:
properties of the component polymers, interaction and adhesion between phases and
morphology(Utracki, 1998). The properties of multiphase materials are strongly dependent on
their phase morphology. Controlling parameters on morphology of polymer blends include
composition of phases, kinematic or rheological properties (elasticity and viscosity), processing
conditions (shear stress, processing temperature,…), and the most important one which is a
thermodynamic parameter: the interfacial tension of the pair of phases(Favis, 2000; Steinmann,
Gronski, & Friedrich, 2001). Thermodynamics of interfaces defines the interfacial tension
between two phases as the area derivative of the surface free energy per unit area of an interface.
Equation 2-11.
Equation 2-12.
⎛ G ⎞
⎜
∂
γ ⎟
ij = and
⎜ A ⎟
⎝ ∂ ij ⎠
⎛ F ⎞
⎜
∂
γ ⎟
ij =
⎜ A ⎟
⎝ ∂ ij ⎠
T , P,
n
T , V , n
where G is the Gibbs free energy, F is the Helmholtz free energy of the whole system, and Aij is
the interfacial area between phases i and j.
There are different ways to measure interfacial tension of two polymers, including theoretical
methods such as Antonoff’s rule(Antonoff, 1942; Antonoff, 1907), the Good and Girifalco
equation(Girifalco & Good, 1957; Good & Girifalco, 1964), and harmonic mean equation, and
experimental methods such as the breaking thread, pendant drop, and sessile drop methods.
12

Antonoff’s rule (Equation 2-13) is valid only when phase 1 and phase 2 have a zero contact
angle and two phases are in equilibrium vapor adsorption(Donahue & Bartell, 1952).
γ = γ − γ
Equation 2-13. 12 2 1
γ ≥ γ
Since polymers have negligible vapor pressure, equilibrium vapor adsorption can not occur
within the experiment time. Therefore, this relation is invalid for polymers.
Good and Girifalco(Girifalco & Good, 1957; Good & Girifalco, 1964) introduced an interaction
parameter(φ) based on work of adhesion as follows:
= 1
W − + γ γ
Equation 2-14 12 2
a
Equation 2-15 ( ) 5 . 0
γ = γ + γ − φ γ
γ
12 2 1 2 1γ
2
The calculated φ values for small-molecule organic liquids give reasonable predictions of the
interfacial tensions(Good & Girifalco, 1964).
The harmonic mean equation(Equation 2-16) is a result of the fractional polarity theory in which
various molecular forces are linearly additive(Wu, 1982). It is valid for low energy
materials(Wu, 1971) and represents the combination of the polar component and the dispersion
component of the work of adhesion as,
γ
Equation 2-16 12 2 1 d d p p
1
d
1
d
2
γ γ
= γ + γ − 4
γ + γ
2
1
2
p
1
1
p
2
γ γ
− 4
γ + γ
Interfacial tension between a low-energy material and a high-energy material is obtained by the
geometric mean equation(Wu, 1971) as follows:
Equation 2-17 ( ) ( ) 5 . 0
d d 0.
5 p p
γ = γ + γ − 2 γ γ − 2 γ γ
12
2
1
Due to high molecular weight and viscosity of polymers, several interfacial tension measurement
methods have been suggested. Generally, calculations of interfacial tensions in these methods are
1
2
1
2
2
13

carried out by comparing minimization of interfacial area and a driving force (gravitational,
centrifugal, shearing force, etc).
Bashforth and Adams(Bashforth & Adams, 1883) related the surface tension to the difference of
density between two liquids and the geometrical profile of a drop. The pendant drop method
determines the profile of a drop of one denser liquid suspended in a less dense liquid at
mechanical equilibrium.
A similar method to pendant drop method is the sessile drop method. In this technique,
interfacial tension is calculated for a liquid surrounded by another liquid with smaller density on
a flat surface based on the balance between gravity and surface forces. It implies that from the
shape of the drop at mechanical equilibrium, interfacial tension can be calculated.
The breaking thread technique is a well-known method in which interfacial tension is calculated
based on the evolution of the shape of a thread imbedded in a second phase. Small distortions
with certain wavelengths are generated at the interface of the phases due to Brownian motion.
These small distortions are expanded based on the interfacial tension and the viscosity of the
polymers. Two theories relate the evolution of the thread to interfacial tension: the theory of
Tomotika(Tomotika, 1935b) and the theory of Tjahjadi(Tjahjadi, Ottino, & Stone, 1994).
Different morphologies can be obtained for melt processing of immiscible polymer blends.
Simple morphologies such as matrix/dispersed particles structures, matrix-fiber structures,
lamellar structures and complex morphologies such as co-continuous structures, composite
droplet structures, and multiple percolated structures can be generated(Favis, 2000; Potschke &
Paul, 2003). These kinds of morphologies can easily be transformed by changing effective
parameters such as applied shear, elongation stress, interfacial tension between phases, and
coalescence and breakup of the phases during processing. For instance, the dispersed phase can
be distorted into long fibrils by applying further shear stress. Breakup mechanism can occur due
to lowering interfacial tension between components dispersing fibrils into droplets. Broken
droplets can be coalesced and form fibrils or larger droplets by coalescence mechanisms. All
other mentioned parameters affect the size and shape of droplets, fibrils, or continuous structure
of phases. The combination of these processes can potentially provide us many kinds of phase
morphologies suitable for various applications.
14

2.1.2.1 Binary Polymer Blends
2.1.2.1.1 Co-continuous Morphology
There is currently an increasing interest in techniques used to control interfacial properties of
polymeric materials. Through controlling morphology of polymer blends, novel materials for
different applications can be designed. For immiscible polymer binary blends, two types of
morphologies, namely matrix/dispersed structure and co-continuous morphologies, have been
realized. At low concentration of one phase, dispersed droplets of that phase in the other
component is observed. By increasing the concentration of the minor phase, phase inversion
occurs and a dual phase region is reached in which various levels of co-continuous structure coexist.
At concentrations near the phase inversion region, blends with two fully interconnected
and interpenetrating structures in three dimensions are observed. In other words, a binary blend
is formed of two polymers in which the surface of each phase is an exact topological replica of
the other (antitropic). Generally, in polymer blends, the matrix dominates the properties of blend;
thus, in binary co-continuous blends both polymers have approximately equal contributions to
the blend properties. Moreover, their interpenetrating assembly can lead to synergistic
improvement of specific conductivity, permeability, and mechanical properties. A process in
which two phases switch their functions is defined as phase inversion. In a certain concentration
range around the phase inversion concentration, which is defined as the co-continuity interval,
co-continuous morphologies are formed. The phase inversion concentration is regarded as the
center of this area and its correct determination is of great importance in which the former
dispersed phase becomes the matrix and vice-versa. Consequently, the position and width of the
phase inversion concentration are influenced by many rheophysical material parameters of the
blend’s constituents, such as interfacial tension. Willemse et al.(Willemse, Posthuma de Boer,
van Dam, & Gotsis, 1999) showed that for different blends with various matrix viscosities and
viscosity ratios, the composition range within which fully co-continuous polymer blend
structures form is influenced by interfacial tension. It was found(He, Bu, & Zeng, 1997) that
phase co-continuity occurs in a wide composition range at a short mixing time, but with
15

increasing mixing time, this interval gets narrow and finally shrinks to one point. Veenstra et
al.(Veenstra, Norder, van Dam, & Posthuma de Boer, 1999) observed a direct correlation
between the width of the co-continuity interval under shear flow and the capillary number. The
above mentioned factors influence the width of the co-phase area, whereas several factors affect
the position of the phase inversion. The most important one is phase viscosities and their ratio.
Investigations have shown that for a binary blend consisting of components of equal viscosity,
phase inversion occurs around a volume fraction of 0.5(Miles & Zurek, 1988b). Several
works(Jordhamo, Manson, & Sperling, 1986; Mekhilef & Verhoogt, 1996) showed that when
component viscosities differ significantly, the phase inversion point is shifted towards
compositions richer in the high viscosity component. Several authors have proposed semiempirical
equations based on rheological properties, particularly viscosity, in order to predict cocontinuity
point. Paul and Barlow(Paul & Barlow, 1980b), according to the observations made
by Avgeropolous(Avgeropoulos, Weissert, Biddison, & Boehm, 1976) and the work developed
by Jordhamo et al.(Jordhamo, et al., 1986), suggested a model for blends prepared at low shear
rates as follows:
Equation 2-18.
[
φ I 1 =
whereφ and η are volume fraction and zero shear viscosity of each phase, respectively, at the
I
phase inversion point. Miles et al.(Miles & Zurek, 1988a) argued that the point of phase
inversion should be more precisely related to the effective viscosity ratio.
φ
I 2
η
η
Equation 2-19. φ η(
& γ )
I 1
1
=
φ η(
& γ ) 2
Most of the researchers have acknowledged that Equation 2-18 and Equation 2-19 are valid for
predicting the phase inversion concentration in many binary blend systems with a viscosity ratio
near unity(Mekhilef & Verhoogt, 1996; Miles & Zurek, 1988a). It has been demonstrated that
they are invalid for the ones that exhibit viscoelastic asymmetry between the melt
components(Favis & Chalifoux, 1988).
I 2
01
0 2
16

Some other studies(Everaert, Aerts, & Groeninckx, 1999; Ho, Wu, & Su, 1990) developed
Equation 2-18 and added excess factors for given blend systems to correlate their results with a
semi-empirical model. A model was derived by Metelkin et al.(Metelkin & Blekht, 1984) based
on theory of filament instability described by Tomotika (Tomotika, 1935a) in terms of viscosity
ratio:
Equation 2-20.
where λ is viscosity ratio and f(λ) is defined as :
1
φ = I 2
1+
λf
( λ)
2
Equation 2-21. f ( λ ) = 1.
25 log( λ ) + 1.
81(log(
λ ))
with λ being the viscosity ratio of the blend components at the blending shear rate. Again all of
these equations can predict the phase inversion point with a very good accuracy when both
components have approximately equal viscosity. Results of samples with high viscosity ratio
demonstrate that these models are unable to predict the phase inversion point and significant
errors appear. For such cases, some authors used the percolation threshold definition to develop
an equation based on maximum packing volume fraction. Utracki(Utracki, 1991) proposed a
model based on theory for mono-dispersed hard spheres(Krieger & Dougherty, 1959).
Equation 2-22.
where
1
[ η ] φ
⎛η
⎞ 1
φ + ( 1−
φ ) ⎜ ⎟
m
m
⎝η
2
φ =
⎠
I 2
1
[ η ] φm
⎛η
⎞ 1 ⎜ ⎟ + 1
⎝η
2 ⎠
φ is the phase inversion point, φ
I 2
m is the maximum packing volume fraction, [η] is
intrinsic viscosity, and indices refer to components. For polymer blends, φ is defined as
m
φ −
m = 1 φc
, where c
φ stands for the percolation threshold and amounts to 0.156 for dispersions of
spheres. The definition of the percolation threshold is explained in detail in the next sections.
This equation explains that the addition of the first component to second one leads to an increase
m
17

in the respective blend viscosity until, at the phase inversion point, the maximum viscosity is
reached.
Luciani et al.(Luciani & Jarrin, 1996), based on Tomotika’s theory and the relative stability of
networks created by coalescence, developed the following equation:
Equation 2-23.
φI 2
2
⎛η
⎞ ⎛ ⎞
1 2 η1
⎜
⎟ Ω ⎜
⎟
⎝η2
⎠ ⎝η2
=
⎠
⎛ ⎞
2
⎜ ⎟
⎛η
⎞ ⎛ ⎞
1 2 η1
2
Ω + Ω ⎜ 1 ⎟
⎜
⎟
⎜
⎟
⎝η
⎠ ⎝ ⎠ ⎜ η1
⎟
2 η2
⎜ ⎟
⎝ η2
⎠
where ⎟ ⎛η
1 ⎞
Ω ⎜ , λ is a complex function of both the viscosity ratio and observed wavelength of the
⎝η
2 ⎠
distortion λ in Tomotika’s equation.
The other semi-empirical model which accounts for the viscous properties of the polymers was
proposed by Willemse et al.(Willemse, de Boer, van Dam, & Gotsis, 1998) based on geometrical
requirements for the formation of co-continuous structures:
Equation 2-24.
1
⎛η
& mγ
⎞
= 1.
38 + 0.
0213⎜
R0
⎟
φ
⎝ σ ⎠
d
where ηm is matrix phase viscosity, R0 is the initial radius of the droplet before deformation into
a cylinder, and σ is interfacial tension. This equation can determine the composition range over
which co-continuity exists and gives the lower limit of continuity for component 1 and upper
limit for component 2.
Steinmann et al.(Steinmann, Gronski, & Friedrich, 2002) derived the following equation point,
assuming that phase inversion occurs at maximum shape relaxation times of domains of
components:
Equation 2-25.
1
1
1 ⎞ z
φ
2I
=
⎛η
⎜
⎟
⎝η
2 ⎠
+ 1
4.
2
18

where the parameter z is dependent on the blend system and is evaluated experimentally for each
system.
Some other researchers focused on the elasticity or used it along with viscosity to predict the
phase inversion composition. The first approach based on the elasticity ratio was described by
Van Oene(Vanoene, 1972). He defined an equation to describe interfacial tension under dynamic
conditions adding an elastic properties term to interfacial tension under static conditions:
Equation 2-26. γ = γ + [ ( N ) − ( N ) ]
where γ eff is the effective interfacial tension under dynamic conditions, d represents droplet
diameter, γ is the static interfacial tension and (N2)d , (N2)m are the second normal stress
functions of the dispersed and matrix phases, respectively.
Considering that the storage modulus G′ and tan δ represent the elasticity of phase i, Bourry and
i
Favis(Bourry & Favis, 1998a) introduced elasticity into the understanding of phase inversion.
eff
d
12
Equation 2-27. φ G′
( ω)
I 1 2 =
φ G′
( ω)
Equation 2-28. φ tanδ
( ω)
I 1
1 =
φ tanδ
( ω)
G ′′ i
whereδ
= . It was reported(Bourry & Favis, 1998a) that these formulas are in much better
′
i
Gi
agreement with experimental data than other equations based only on viscous effects,
particularly at high shear rates. Steinmann et al.(Steinmann, et al., 2001) compared some of the
mentioned approaches and found that the factors are still inadequate to predict the phase
inversion point. Consequently, they developed a semi-empirical equation based on elasticity ratio
Ψ eff . G′′
:
Equation 2-29. φ
− . 34log(
Ψ ) + 0.
67
I 2
I 2
= 0 I 1
eff . G′
′
1
2
2
d
2
m
19

As reviewed above, numerous models have been proposed but to date none of them are able to
predict the phase inversion composition for all binary polymer blends.
One of the most important applications of co-continuous blends is the fabrication of continuous
conductive parts in the blend. Several works have been done in this area to decrease the
percolation threshold of conductive polymers by making a network of conductive polymer in
another phase. As conductivity of the blend corresponds to the continuity of the conductive
polymer, the percolation threshold of continuity of the conductive component can be measured
by showing the percolation threshold of conductivity of the blend. Zilberman et al.(Zilberman,
Siegmann, & Narkis, 2000d) compared the conductivity of several binary blends such as
PANI/LLDPE, PANI/CoPA and PANI/(PS/DOP) (Figure 2-1). A detailed study of binary
conductive polymer blends will be conducted in the next sections.
Figure 2-1. Electrical conductivity versus PANI content of polymer/PANI binary
blends(Zilberman, et al., 2000d).
2.1.2.1.2 Interfacial Modification of Binary Blends
Compatibilizing agents (surfactants) have been long used with mixtures of low molecular weight
molecules. The earliest and simplest kind is soap made by boiling fats together with plant ashes,
which dates back to 2800 B.C. and was industrially developed in 1920’s. The term of
20

“compatibilization” was suggested by Gaylord in the 1960s for the method used to decrease the
interfacial tension between polymer pairs and the blend shows more stable morphology(N. G.
Gaylord, 1969; N. G. Gaylord, ,Vol.142,P.76,(ACS,Washington,1975). 1975).
Compatibility between phases is enhanced using a block copolymer with particular physically or
chemically compatible segments with each of the blend components. Block copolymers are
designed for binding two components together. They stay at the interface and either part of their
molecular chain has physical or chemical interaction with each component of the blend. The
main effect of addition of an interfacial agent on the morphology of an immiscible blend is to
reduce the particle size and to narrow the particle-size distribution(Hoganand & Banks, 1958;
Matos, Favis, & Lomellini, 1995; Sarazin & Favis, 2003) (Figure 2-2).
a) b) c)
d) e) f)
Figure 2-2. Action of the diblock copolymer to reduce the pore size for 50/50 PLLA/PS
blends at copolymer concentrations of (a) 2.5%, (b) 5%, (c) 8%, (d) 12%, (e) 15%, and (f) 20%.
Copolymer concentration shown is based on the weight of polystyrene. The white bar indicates 1
micron(Sarazin & Favis, 2003).
Both reduction of coalescence and decrease in interfacial tension results in reduction of particle
size after interfacial modification. Sundararaj et al.(Sundararaj & Macosko, 1995) showed
suppression of coalescence happens when two droplets with a layer of diblock or graft
copolymer at the interface are less likely to coalesce due to formation of a shell of copolymer
molecule around droplets.
21

2.1.2.2 Ternary Polymer Blends
Mathematical models derived from physical considerations have been successfully used to
predict and explain several observed morphologies of immiscible ternary polymer blends. Harkin
used thermodynamic properties, such as surface and interfacial tensions of the components, to
express the tendency of liquid to spontaneously spread across a solid or liquid substrate(Harkins
& Feldman, 1922). The first comprehensive study on a ternary liquid mixture was carried out by
Torza and Mason(Torza & Mason, 1970). They used modified Harkins Spreading Theory to
analyze interfacial energy differences between the liquid components. Hobbs et al.(Hobbs,
Dekkers, & Watkins, 1988) followed their work and employed this method to predict and
interpret composite-droplet morphology of ternary polymer blends. The modified Harkins
equation can be written as:
Equation 2-30. λij = γ jk − γ ik − γ ij
λ is defined as the spreading coefficient giving the tendency of component (i) to encapsulate or
ij
spread onto component (j) in the matrix of component (k). γ , ij γ , and jk γ are the interfacial
ik
tensions of the different polymer pairs. Two different morphologies can be obtained, depending
on the values of the spreading coefficients, as spreading is predicted to occur only for positive
values ofλ . In other words, a positive value of one of the spreading coefficients such asλ ij
ij
indicates wetting of phase j by phase i. This case is called complete wetting morphology and is
shown in Figure 2-3. The other possible prediction is partial wetting in which all spreading
coefficients have negative values. In this morphology all phases have interfaces with each other
(Figure 2-3).
22

a)
λ BC
A and B are matrices,
b)
C is minor phase
A and C are matrices,
c)
B is minor phase
B and C are matrices,
A is minor phase
<
0
Figure 2-3. a), b), c) Ternary blends with partial wetting case, and d), e), f) ternary blends with
complete wetting case
A phenomenological model suggested by Guo et al.(Guo, Gvozdic, & Meier, 1997; Guo,
Packirisamy, Gvozdic, & Meier, 1997) was used to predict the morphology of ternary and
quaternary blends. They used interfacial area associated with interfacial tension and derived an
expression for the free energy of mixing of a multiphase blend. The free energy calculation is
written as:
Equation 2-31. G = ∑ni
μi
+ ∑ Aiγ
ij
i
d)
e)
f)
i≠
j
λ BC
A is matrix, B and C
are minor phases
B is matrix, A and C
are minor phases
C is matrix, A and B
are minor phases
>
0
23

where μ is the chemical potential of component i, ni is the number of moles, Ai is the interfacial
i
area, and γ ij is the interfacial tension between components i and j. Thus, the phase structure of a
polymer blend can be predicted by comparing the Gibbs free energies of the different structures.
In their work on a HDPE/PS/PMMA blend, numerous conditions were assumed to simplify the
study such as equal value of the first term of the equation for all types of morphology. For
further simplification, they assumed that numbers of droplets of minor phases were equal to each
other, which lead to a big error.
The second term is simplified further and rewritten as:
Equation 2-32.
Equation 2-33.
Equation 2-34.
1 1 2
1
3
( ∑ Aγ
) = ( 4π
) 3 ⎡ n 3 x γ + n 3γ
⎤(
3V
)
i
1 1
2
1
3
( ∑ Aγ
) = ( 4π
) 3 ⎡ n 3 ( 1+
x)
γ + n 3γ
⎤(
3V
)
i
( ) ( ) ( ) 3
1 1 2
1
2
2
3 3 3
3
3
∑ Aγ
= 4π
⎡ n x γ + n ( 1+
x)
γ ⎤ 3V
i
ij
ij
ij
B+
C
B
C
C
B
where Vi is the volume fraction of phase I,
V
V
C
⎢⎣
⎢⎣
⎢⎣
B
B
B
AB
BC
AB
C
C
AC
C
⎥⎦
BC
c
⎥⎦
2
3
AC
⎥⎦
c
24
B x = , nB and nC are numbers of particles of B and
C phases. The most stable morphology is predicted by calculation of the values of ( ∑ Aγ )
which correspond to the value of the Gibbs energy of mixing.
Obviously, the Harkins equation and the Guo model are purely thermodynamic equations in
which some other factors influencing the morphology, such as rheological parameters, have not
been considered. Although in measuring interfacial tensions of polymer pairs by various methods
such as the breaking thread and pendent drop, viscosities of the components are taken into
account, the effect of elasticity is still neglected. In the method developed by Guo to calculate the
free energy compared to Harkins equation, concentration of components is also contributed to
determine the kind of morphology as well and it is more modified. Researchers use both the
spreading coefficients and the minimal free energy surface model to predict the morphologies of
polymer blends (Guo, Gvozdic, et al., 1997; Guo, Packirisamy, et al., 1997; Hara & Sauer, 1998;
Hobbs, et al., 1988; Horiuchi, Matchariyakul, Yase, & Kitano, 1997; Nauman & He, 2001;
i
2
3
ij
c

Reignier & Favis, 2000, 2003a; Reignier, Favis, & Heuzey, 2003; Tchomakov, Favis, Huneault,
Champagne, & Tofan, 2004; Urashita, Kawakatsu, & Doi, 2000).
It has been observed that the rheological properties of the two minor phases can play a role in the
final morphology of ternary polymer blends, most notably through viscosity effects(Gupta, 1993;
Luzinov, Pagnoulle, & Jerome, 2000; Nemirovski, Siegmann, & Narkis, 1995; Tchomakov, et
al., 2004), and elasticity ratio effects(Legros, Carreau, Favis, & Michel, 1997; Vanoene, 1972).
The effect of the viscosity ratio is still controversial amongst the authors. Some authors did not
observe any influence of the viscosity ratio on the type of morphology(Reignier, et al., 2003;
Tchomakov, et al., 2004), while some observed encapsulation of the higher viscosity component
by the component of lower viscosity(Nemirovski, et al., 1995), and others observed a contrary
result(Gupta, 1993).
Van Oene(Vanoene, 1972) showed that under conditions of dynamic flow, the difference in
elasticity between the phases can contribute to the interfacial tension. A dynamic interfacial
tension, which can be quite different from the static interfacial tension, can have a significant
impact on the final morphology. He developed an expression for the interfacial tension in flow
and defined a term proportional to the difference between a second normal stress function in this
equation(Equation 26)
2.1.2.2.1 Composite Droplet Morphology
Another interesting morphology observed in multiphase polymer blends is the composite
droplets morphology. In this particular morphology, a dispersed phase (B), in a matrix of (A),
encapsulates a third phase (C).
There are two main causes for sub-inclusion formation: 1) Thermodynamic effect: to minimize
the surface free energy of a blend (Guo, Packirisamy, et al., 1997; Hobbs, et al., 1988) and 2)
Rheological effect: the entrapment of the sub-inclusions in a more viscous phase near the phase
inversion region(Favis & Chalifoux, 1988).
Reignier et al.(Reignier & Favis, 2000, 2003a, 2003b; Reignier, et al., 2003) showed that in a
ternary blend comprised of PMMA, PS, and HDPE, when HDPE is the matrix, the blend exhibits
25

composite droplet formation constituted of a well-segregated PMMA core and PS shell (Figure
2-4Ia, 2-4Ib, and 2-4Ic).
I) II)
a)
b) c)
Figure 2-4. SEM micrographs of the 80 (HDPE) /20 (PS/PMMA) I) a) fracture surface etched
with acetic acid for 2 min, 14%PS/86%H-PMMA; b) microtomed surface etched 24 h with
cyclohexane, 9%PS/91%H-PMMA; and c) microtomed surface etched 24 h with cyclohexane,
9%PS/91%L-PMMA, II) Evolution of the shell formation process with increasing PS content
(vol. % based on the dispersed phase) for the 80(HDPE)/20(PS/L-PMMA)(cryofractured
samples after being etched for 10 s with acetic acid).
It was explained that the thickness of the PS shell can be controlled by changing the relative
amount of PS and PMMA. They also showed the evolution of the shell formation process with
increasing PS content in a ternary HDPE/PS/PMMA blend(Reignier & Favis, 2003a) (Figure 2-
4IIa, 2-4IIb, and 2-4IIc). They also found(Reignier & Favis, 2000) that after two minutes of
mixing, approximately all of the PMMA is encapsulated within the PS dispersed phase and
morphology becomes stabilized after that.
Harrats et al.(Harrats, Omonov, & Groeninckx, 2005) reported encapsulation of the PP phase
between the PS and PA6 phases (Figures 2-5a and 2-5b). They found this against the prediction
of the Harkins equation, whereas the interfacial tension values of γPA6/PS=13.72 mN/m,
γPA6/PP=13.64 mN/m, and γPP/PS=2.26 mN/m demonstrated partial wetting case. They did not
a)
b)
c)
26

measure the interfacial tension of tensions. The harmonic mean equation, as mentioned earlier,
is a method to calculate the interfacial tension between two polymers based on their surface
tension and polarity (Equation 16).
a) b)
Figure 2-5. SEM micrograph of 70PA6/15PP/15PS blends: (a) cryofractured and (b) chloroform
extracted surfaces.(Harrats, et al., 2005)
They claimed that the concept of spreading theory fails in their case because “the development of
the phase morphologies in polymer melt depends also on other key factors such as the viscosity
and elasticity of the blend. The use of the interfacial tension alone for the prediction of the type
of phase morphologies in immiscible polymer melt can be successful in blend components
having closer melt viscosities.”
27

Reignier et al.(Reignier, et al., 2003) studied the effect of different factors on the composite
droplet formation. They showed the influence of elasticity on composite droplet formation by
introducing the dynamic interfacial tension term taken from VanOene(Vanoene, 1972) in the free
energy model developed by Guo et al.(Guo, Packirisamy, et al., 1997), and developed new
equations for the interfacial free energies in conditions of dynamic mixing:
Equation 2-35.
Equation 2-36.
Equation 2-37.
2 ⎡ Ri
⎤ 2 ⎡ Ri
( ∑ Aγ
) = 4πR
γ + ( N − N ) + 4πR
γ + ( N − N )
i
⎢
⎣
2 ⎡ Re
⎤ 2 ⎡ Ri
( ∑ Aγ
) = 4πR
γ + ( N − N ) + 4πR
γ + ( N − N )
i
ij B+
C
ij B
C
⎢
⎣
2 ⎡ Re
⎤ 2 ⎡ Ri
⎤
( ∑ Aiγ
ij ) C = 4πRe
⎢γ
CA + ( N1,
C − N1,
A ) ⎥
+ 4πRi
⎢γ
BC + ( N1,
B − N1,
C ) ⎥⎦
B
e
i
⎣
BA
BA
Normal stress differences for the phases A, B, and C are referred to by N1, γ ij is the interfacial
tension between the components i and j, and the internal and external radius of the core-shell
droplets are shown by Ri and Re, respectively. Reignier et al.(Reignier, et al., 2003) indicated that
the viscosity ratio estimated at a constant shear stress rather than at a constant shear rate is
accurate to study the effect of viscosity ratio on the morphology because shear stress is more
continuous at the interface between the dispersed phase and the matrix phase.
The effect of molecular weight of phases on the encapsulation process was
investigated(Reignier, et al., 2003). The equilibrium morphology for HDPE/low molecular
weight PS/low molecular weight PMMA illustrates that PS encapsulates PMMA (Figure 2-6a).
When components were replaced by high molecular weight PS and high molecular weight
PMMA, an inversion in encapsulation occurred and PMMA encapsulated PS (Figure 2-6b). They
related this discrepancy to the high viscosity of PS that does not allow PMMA to be
encapsulated. Finally, considering all the observations, they concluded that interfacial energy
reduction is the main driving force controlling encapsulation effects and that the viscosity ratio
and the absolute viscosity have little influence on encapsulation phenomena in composite
droplets.
6
6
6
1,
B
1,
B
1,
A
1,
A
⎥
⎦
⎥
⎦
⎦
i
i
⎢
⎣
⎢
⎣
⎣
CA
CB
6
6
6
1,
C
1,
C
1,
A
1,
B
28
⎤
⎥
⎦
⎤
⎥
⎦

a) b)
Figure 2-6. Dependence of the composite dispersed morphology on PS and PMMA molecular
weights. SEM photomicrographs of (a) HDPE/L-PS/L-PMMA; PMMA is extracted by acetic
acid and (b) HDPE/H-PS/H-PMMA; PS is extracted by cyclohexane. The white bar denotes
1 μm(Reignier, et al., 2003).
This section showed that many parameters can affect the simplest morphology of ternary
polymer blends, which is composite droplet morphology. The question is: which theory can
correctly predict the morphology, and which theory can give similar results to reality in the
fastest time? In some work, different theories mentioned above are compared by (Valera, Morita,
& Demarquette, 2006) work. In this work, the morphology of the PMMA/PP/PS ternary blend
with different compositions was comprehensively studied. PMMA, PS, and PP are selected as
the matrix and the composition of the minor phases was changed to find which model is able to
predict a wide range of compositions. It has been reported that the predictions of the minimal
surface free energy theory were verified by experimental observations only when PS was the
matrix, and in the other cases the minimal surface free energy theory was unsuccessful to predict
the morphology. The spreading coefficient model and the dynamic interfacial tension model can
predict most of the experimental results.
Finally, it is concluded that the fastest way to predict the morphology of the ternary polymer
blend with a great accuracy still remains the spreading coefficients.
29

2.1.2.2.2 Double-Percolated Systems
Zhang et al.(J. Zhang, et al., 2007) showed that through a combination of composite droplet and
co-continuous blend morphology preparation methodologies, a continuous shell structure of one
component was situated at the interface of a co-continuous structure of two other phases. In a
composite-droplet structure with composite droplets of PMMA in PS, if the amount of HDPE is
decreased while the quantity of the PMMA is increased, a co-continuous network of HDPE and
PMMA is formed while the PS forms a continuous sheath structure at the HPDE/PMMA
interface (Figure 2-7a). In this kind of morphology, it is possible to have a co-continuous
structure of one component using a very small amount of that component. The percolation
threshold is significantly decreased and continuity development increases as compared to
classical blends. This is verified by gravimetric measurement, as shown in Figure 2-7b. As it is
illustrated, the percolation threshold of the PS phase in a HDPE/PS blend and that of the PMMA
phase in a HDPE/PMMA blend are 15% and 20%, respectively. The percolation threshold of PS
sharply decreases when it is situated at the interface of HDPE and PMMA, whereas at
concentrations as low as 3%, 70% continuity is observed.
a) b) c)
Figure 2-7. a, b) FIB-AFM images of HDPE/PS/PMMA ternary blends. The white line in Figure
7b indicates the section analyzed below, and c) Continuity of PS or PMMA as a function of the
composition using the solvent dissolution technique(Zhang, et al., 2007)
30

Omonov et al. found double percolated structures for ternary 40PA6/30PP/30PS (Figure 2-8a)
and 40PA6/(25/5)(PP/PP–MA2)/(25/5)(PS/SMA2) (Figure 2-8b) blends. In this work, the
observed results were opposite to the prediction of spreading theory, which is partial wetting.
They found that despite the negative values of spreading coefficients, a complete wetting case
with PA6 phase at the interface is observed. The gravimetric measurements for all phases
showed that for 40PA6/30PP/30PS, all phases have more than 97% continuity while PA6
separates the PP and PS phases. Figure 2-8b shows that addition of a compatibilizing agent to the
sample does not change the location of phases. It only reduces phase sizes, but PA6 still remains
at the interface of the two other phases.
a) b)
Figure 2-8. Scanning electron micrograph of a) 40PA6/30PP/30PS; formic acid extracted
surfaces, and b) 40PA6/(25/5)(PP/PP–MA2)/(25/5)(PS/SMA2); formic acid extracted surfaces
(Harrats, et al., 2005)
One of the most important applications of double-percolated structures is the preparation of
conductive polymer blends. Since locating the conductive phase at the interface decreases the
percolation threshold of continuity of the middle phase, it is predicted that the percolation
threshold of conductivity of the blend is reduced significantly. Narkis et al.(Narkis et al., 2000b;
Zilberman, Siegmann, & Narkis, 1998, 2000a, 2000b) studied various polymer blends consisting
of two immiscible thermoplastic polymers and PANI and tried to locate PANI at the interface of
two other polymers, but they were not successful. They showed that for CoPA/LLDPE/PANI
(Figure 2-9) and (PS/DOP)/LLDPE/PANI (Figure 2-10), changing the concentration of the
31

middle phase has an effect on the continuity and conductivity of the PANI, which is the inner
phase. It is noted that as PANI is not the middle phase, the percolation threshold is not reached at
lower concentrations of PANI such as 10%, and a higher amount of PANI is required(Zilberman,
et al., 2000b). A detailed review on the double percolated morphologies is presented in the next
sections.
Figure 2-9. Electrical conductivity versus LLDPE content for CoPA/LLDPE/PANI ternary
blends containing 10 and 20 wt% PANI(Zilberman, et al., 2000b)
Figure 2-10. Electrical conductivity versus LLDPE content for (PS/DOP)/LLDPE/PANI blends
containing 10 and 20 wt% PANI(Zilberman, et al., 2000b)
32

2.1.2.2.3 Effect of Composition on Ternary Blends with Complete Wetting
Morphology
In the case of complete wetting, one phase situates at the interface of two other phases and
separates them from each other. By varying the composition, three different cases can be formed.
Guo et al.(Guo, Packirisamy, et al., 1997) showed that in a HDPE/PS/PMMA ternary blend, PS
always separates HDPE and PMMA phases for various compositions. They showed that when
HDPE is the matrix and PS and PMMA are the minor components, the PMMA phase is
encapsulated by the PS phase (Figure 2-11a).
a) b)
c)
Figure 2-11. a) 70%HDPE/15%PS/15%PMMA, b) 15%PMMA/70%PS/15%HDPE(PMMA
extracted fracture surface), and c) 70%PMMA/15%PS/15%HDPE (PS extracted fracture surface)
(Guo, Packirisamy, et al., 1997)
33

When PS is the matrix with PMMA and HDPE minor phases, separate dispersions of HDPE and
PMMA have the lowest interfacial free energy (Figure 2-11b). Finally, when PMMA is the major
phase and PS and PMMA are minor components, the HDPE is encapsulated in the PS phase
(Figure 2-11c).
2.1.2.2.4 Interfacially Modified Ternary Polymer Blends
As mentioned above, the Harkins theory can predict whether the morphology of a ternary blend
is complete wetting or partial wetting. This theory is based on the interfacial tension of polymer
pairs. Hence, by changing the interfacial tension between two polymers via addition of
interfacial modifier, the prediction of the morphology by Harkins can toggle between the
complete wetting case and the partial wetting case. Another effect of interfacial modifier is that
in a partial wetting case, one phase moves from one phase to the other phase, thus creating more
interfacial area. Guo et al. (Guo, Packirisamy, et al., 1997) showed that in a ternary polymer
blend of 70/10/20 HDPE/PS/PP, all three phases have interface with each other (Figure 2-12a).
This is predicted by the Harkins equation, as interfacial tension values of γPE/PS=5.9 mN/m,
γPE/PP=1.1 mN/m, and γPP/PS=5.1 mN/m show three negative spreading coefficients. A blend
without interfacial modifier represents more interfacial area between the PS phase and the PP
phase (Figure 2-12a). Addition of 0.5% PS/PE diblock copolymer shows movement of the major
part of the PS phases to HDPE (complete wetting case), and also more PS droplets at the
interface move to the HDPE phase (Figure 2-12b).
The other effect of interfacial modifier on the blend, as previously discussed, is increasing
interfacial area between phases. Therefore, addition of diblock copolymer to the blend results in
a decrease of the droplet size. Figure 2-12c shows that addition of 2% PS/PE copolymer to
70/10/20 HDPE/PS/PP results in smaller droplet sizes in both the partial and complete wetting
cases.
34

a) b)
C
B
A
c)
Figure 2-12. SEM photomicrograph of a) 70%HDPE(A)/20%PP(B)/10%PS(C), b)
70%HDPE(A)/20%PP(B)/10%PS(C) + 0.5% S-E block copolymer, and c)
70%HDPE(A)/20%PP(B)/10%PS(C) + 2% S-E block copolymer(Guo, Packirisamy, et al., 1997)
2.1.2.3 Quaternary Polymer Blend Systems
A
C
A few works have been performed to predict and analyze the morphology of quaternary blend
systems(Guo, Gvozdic, et al., 1997; Virgilio, Desjardins, et al., 2009). Guo et al.(Guo, Gvozdic,
et al., 1997) calculated the value of for various cases of a quaternary blend comprised of
PP, HDPE, PS, and PMMA. They showed, as predicted when PP is the matrix, that a structure of
(PE+PS/PMMA) is observed where the PMMA phase is encapsulated by the PS phase and the
PE particles are dispersed separately in the matrix (Figure 2-13a).
B
B
A
C
35

a) b)
c) d)
Figure 2-13. SEM photomicrograph of a) 10%PMMA(D)/10%PS(C)/60%PP(A)/20%HDPE(B),
b) 10%PMMA(D)/60%PS(A)/20%PP(B)/10%HDPE(C) , c)
60%PMMA(A)/20%PS(B)/10%PP(C)/10%HDPE(D), and d)
10%PMMA(D)/10%PS(C)/20%PP(B)/60%HDPE(A)(Guo, Gvozdic, et al., 1997)
36

In the case where PS is the matrix, the value of interfacial energy results in a
(PMMA+PP/HDPE) structure, in which HDPE is encapsulated by PP and PMMA is separately
dispersed in the matrix (Figure 2-13b). For PMMA and HDPE matrices, structures of
(PS/PP/HDPE) (Figure 2-13c) and (PP/PS/PMMA) (Figure 2-13d) were predicted and observed,
respectively.
2.1.3 Porous Polymeric Substrates Prepared by Melt-Blending
Highly controlled polymeric co-continuous structures have been fabricated in our group(J. Li &
Favis, 2001b; Jianming Li, Ma, & Favis, 2002; Sarazin & Favis, 2003) through a melt-blending
process of immiscible polymers followed by selective solvent extraction. As mentioned before,
in a binary immiscible polymer blend, two possible morphologies can be formed: dispersed
droplet/matrix and co-continuous morphologies. Binary polymer blends comprised of polymer 1
and polymer 2 in the co-continuity range lead to a co-continuous morphology at concentrations
above the percolation threshold. The dual phase region is located at concentrations above the
percolation threshold, where phase inversion determines the area that two fully interconnected
phases through a continuous pathway exist. Etching one of the components by selective solvent
extraction leads to a fully interconnected porous substrate. The greatest advantage of this method
compared to other techniques used for generating interconnected porous polymer substrates, such
as solvent casting/particulate leaching or gas foaming, is the ability to control porosity and pore
distribution, and internal surface area.
2.1.3.1 Controlling the Pore Size in the Substrate
There are a few ways to control the pore sizes of porous samples prepared via melt-blending.
Pore size and extent of continuity can be controlled through composition, interfacial
modification, and static annealing. Interfacial modification in a blend is employed to achieve
37

substantially reduced pore sizes, and annealing of the multi-component is used to increase the
phase sizes of porous material. It has been observed that for a constant void volume, a range of
pore diameters from 0.9 to 72 μm can be obtained(Sarazin & Favis, 2003).
2.1.3.1.1 Coarsening of Binary and Ternary Polymeric Systems
In some works, large porous media are needed for facilitating penetration of the solution through
the empty interconnected network of a porous substrate. In order to perform controlled
annealing, the sample must be in contact with hot press surfaces for a certain time. In this case, a
temperature gradient, which is defined as the change in temperature over the change in distance
from hot surfaces, is present throughout the samples. At each point in the sample (x,y,z) in time
t, the temperature is T(z,t). By increasing temperature, the magnitude of gradient rises and T
equals T(z,t) until it reaches the set temperature. At this point the gradient is no longer a function
of time and just changes in height. It is concluded that for thick samples(larger than 0.5cm
thickness), the annealing method is not a suitable technique to fabricate porous substrates with
uniform pore sizes.
In a detailed work about annealing of immiscible co-continuous blends, Zhenhua et al.(Zhenhua
Yuan, 2005) proposed that the driving force for the coarsening process during static annealing is
a capillary pressure effect. Siggia(Siggia, 1979) explained that although capillary pressure is
proportional to the interfacial tension, it is inversely proportional to the rod thickness. It was
described that in a blend, thin part structures with small cross-sections generate capillary forces
larger than thick parts with large cross-sections. It demonstrates a force gradient from thin parts
to thick parts and a continuous merging of thin parts toward the thick ones resulted from the
difference in capillary pressure throughout the co-continuous structure (Figure 2-14).
Hence, there are a lot of thin rods or branches attached to thick rods which continuously feed
them. Important factors in coarsening were reported as temperature and time of the annealing,
interfacial tension, and zero shear viscosity of components. Increasing the time and temperature
of annealing results in increasing the droplet motions and coalescence of droplets, which leads to
average domain size growth and also size distribution growth(Willemse, Ramaker, van Dam, &
38

Posthuma de Boer, 1999) (Kumin & Chang Dae, 1996). Linear time dependence was observed in
the coarsening of immiscible co-continuous blends as shown by experimental pore size growth at
a constant temperature of 200°C in Figure 2-15. Sarazin et al.(Sarazin & Favis, 2003; Sarazin,
Roy, & Favis, 2004) showed that for a binary blend of PLLA and two typical polymers, domain
growth is more rapid at a higher temperature over a longer time.
.
a) b)
c) d)
Figure 2-14. Coarsening process in co-continuous structure upon annealing.(a) the thin rod
merges/flows into thick ones at both sides, the thick rod merges into a much thicker one at the
same time; (b) the thin rod becomes thinner during the merging process; (c) the thin rod
breaks/splits; and (d) the residual parts retract/shrink to the thick rods.(Yuan, 2005)
Veenstra et al.(Harm Veenstra, Van Dam, & Posthuma de Boer, 2000) described a relationship
between coarsening and the fiber retraction process. They showed the evolution of a polymer rod
to a sphere in a polymer matrix during the annealing process. However, this mechanism is unable
to explain some facts about coarsening and is not in agreement with experimental data(Zhenhua
Yuan, 2005).
Reignier et al.(Reignier & Favis, 2000) investigated static coalescence for a composite droplet
morphology in a ternary blend. A similar mechanism governed by the same methodology for
coalescence of each phase in binary blend and ternary blend was found. They set the temperature
at 200°C and observed that core-core and shell-shell coalescence occurred from 2 min to 90 min.
They reported that the only difference is that the ternary blend showed a dual coalescence
process, first between subinclusions within composite droplets and second between composite
39

droplets. The results indicate that due to forces and coalescence, each phase joins with same
phase, and thus phase sizes become larger. In other words, a complete transition from multiple
subinclusion particles within a given shell to a core-shell structure is observed. It can be said that
this process is performed similarly to the case of binary blends, as thin parts of each component
are being pushed to thick parts of the same component due to capillary forces, which results in an
increase in average pore size.
Figure 2-15. Experimental pore size growth (points) with the theoretical model (solid lines) for
different blends annealed at 200°C. (Yuan, 2005)
2.2 Percolation Theory
The behavior of the connected clusters in a random graph in mathematics is described by the
percolation theory. The main concept of the percolation theory is the existence of a percolation
threshold. Percolation threshold is a mathematical term defined as the formation of long-range
connectivity in random systems. Percolation threshold in engineering and material science deals
with fluid flow and similar processes which concern the movement and filtering of fluids through
porous media. One question that arises is whether a liquid poured on top of a porous material is
able to make its way from the top to the bottom. From a mathematical point of view, a regular
40

lattice, such as square lattice, with a random network by randomly occupying sites (vertices) or
bonds (edges) with a statistically independent probability is imagined (Figure 2-16).
If it is examined as an infinite network corresponding to Kolmogorov`s zero-one law, for a
given P, the probability that an infinite cluster exists is either zero or one. There is a critical P
(since this probability is increasing) below which the probability is always 0 and above which
the probability is always 1. This critical amount Pc is called the percolation threshold, which if
we start out at P = 0 and randomly create connections, Pc is defined as a point at which a
spanning cluster first appears. It means that for P greater than Pc, a spanning cluster always
exists, although some isolated, non-spanning clusters can be present, and for P less than Pc, there
are only isolated clusters(Stauffer & Aharony, 1994a).
Figure 2-16. Detail of a bond percolation on the square lattice in two dimensions with
percolation probability p=51(Stauffer & Aharony, 1994b)
Pc may be calculated theoretically in some cases such as for a square lattice in two-dimensions
where the percolation threshold has a value of 0.5. For triangular and honeycomb bonds, (Pc)
will be respectively as follows:
⎛ π ⎞
Equation 2-38. Pc = 2Sin⎜
⎟
⎝18⎠
41

⎛ π ⎞
Equation 2-39. Pc = 1−
2Sin⎜
⎟
⎝18⎠
Calculation of Pc for a square lattice was an open question for more than 20 years and was finally
resolved by Harry Kesten. For three-dimensional systems, Pc’s are unknown and cannot be
calculated for now, although the most interesting thing to do is not calculating the Pc.
The percolation threshold has a significant application in universality systems. For example,
Utracki et al.(Utracki, 1991) used the percolation threshold to predict the phase inversion point
based on emulsion theory as follows:
Equation 2-40.
η ⎛φm
φ ⎞
1 − 2 =
η ⎜
φm
φ ⎟
2 ⎝ − 1 ⎠
In this model, m and η refer to the volume fraction of the component at percolation threshold
and intrinsic viscosity, respectively.
The universality principle describes that the value of Pc is connected to the local structure of the
graph, while the behavior of clusters below, at and above Pc are invariants of the local structure,
and therefore, in some sense are more natural quantities to consider.
2.2.1 Universality Systems
Systems which tend to be chaotic often have a large number of interacting parts. They are called
universality systems. Systems display universality in a scaling limit, when a large number of
interacting parts come together. As a matter of fact, universality originated in the study of phase
transitions in a statistical science, especially in statistical mechanics. The transformation of a
thermodynamic system from one phase to another is a phase transition. When the properties of a
material change in a dramatic way, a phase transition happens. For example, when a magnet is
heated, its magnetism is lost. In this case, the phase transition is characterized by the
[ η ] φm
42

magnetization, which is a function of a parameter of the system, such as the temperature. A
system’s critical point is the value of the parameter at which the system has a phase transition.
For a given parameter, such as (φ ), with a critical value of ( φ ), the order parameter (a) is
c
approximated by the following equation:
Equation 2-41. ( ) t
a a φ −φ
= 0
The exponent t is a critical exponent of the system and its value is dependent on the dimensions
of the lattice(Grimmett, 1999). For a monodispersed distribution of spheres, a value of 0.156 is
predicted for onset of the percolation threshold by percolation theory, which has been verified by
several works.
For example, in the case of an electrically conductive polymer with filler, the effect of filler
loading on the composite resistivity follows a nearly universal pattern regardless of which fillers
are chosen.
2.3 Conductive Materials
In modern society, a life without electronics cannot be imagined. All materials under normal
conditions have some resistance to flowing charges. The electrical resistance of a material is
defined by Ohm’s law as the ratio of the voltage (V) applied to the electric current (I) which
flows through it. A resistance formula is:
There are three main groups of materials from an electrical point of view: conductors,
semiconductors, and insulators. Materials containing movable charges of electricity are called
conductors. These mobile charges are forced to move when an electric potential difference is
applied across separate points on a conductor. The measure of the ability of a conductor to carry
electric current is conductivity, which is defined as the ratio of the amount of charges passing
through a unit area of the conductor per second, divided by the electric field intensity.
c
43

Equivalently, it is the inverse of the electrical resistivity, with units in Siemens per meter.
Therefore, it can be written as:
Equation 2-42
where R is resistance, is resistivity, l is length, and A is cross sectional area.
Devices made from semiconductor materials are the foundation of modern electronics, including
radio, computers, telephones, and many other devices. Semiconductor devices include
transistors, solar cells, many kinds of diodes including the light-emitting diode, the silicon
controlled rectifier, and digital and analog integrated circuits. In semiconductors, electrical
conductivity is controlled over a wide range, either dynamically or permanently. These materials
generally have an intermediate conductivity (Figure 2-17) which varies under different
conditions such as specific frequencies in AC, light, temperature and electric field. Insulators are
non-conducting materials as they lack mobile charges.
Figure 2-17. Logarithmic conductivity ladder locating some metals and conducting
polymers(Pratt, 2001)
44

Polymers have long been thought of and applied as insulators. Moreover, some observed
electrical conduction in polymers was generally regarded as an undesirable phenomenon. With
the discovery in 1960 of intrinsically conducting polymers (ICPs), a new class of electronically
conducting polymers in the partially oxidized state was surprisingly discovered. In metals and
semiconductors such as silicon, free electrons and holes are the charges that are responsible for
the current. The Nobel Prize in Chemistry in 2000 “for the discovery and development of
electronically conductive polymers” was received by Alan J. Heeger, Alan J. MacDiarmid and
Hideki Shirakawa(Chiang, Druy, et al., 1978; Chiang et al., 1977; Chiang, Park, et al., 1978).
2.3.1 Electrically Conductive Polymers and Composites
Generally, four types of electrically conductive polymer system with different degrees of
conductivity have been recognized(Wise, Wnek, Trantolo, Cooper, & Gresser, 1997).
1. The most widely used conducting polymeric systems where an insulating polymer matrix
is filled with particulate or fibrous conductive fillers such as metal powder, carbon black, or
carbon nanotubes to impart high conductivity.
2. Ionically conducting polymers are electrically conductive because of the movement of
ions present in the system. For example mobility of lithium ions in polyethylene oxide is behind
the electrical conductivity in this system.
3. Redox polymers containing immobilized redox centers (electroactive centers) can
conduct charge by electron transfer from one center to another through the well known
“hopping” mechanism. During conduction, electrons tunnel from one redox center to another
through an insulating barrier. The systems need to have a large number of redox centers to
increase the probability of such tunneling.
4. Intrinsically conducting polymers are conjugated polymers consisting of alternating
single and double bonds, creating an extended π-network. The movement of electrons within this
π-framework is the source of conductivity. However, dopant is required to increase the level of
conductivity for this type of polymers.
45

2.3.1.1 Conducting Polymer Composite Materials (CPCM)
The development of efficient conductive polymer composites materials(CPCMs) remains an
important endeavor in light of growing energy concerns. Conductive and anti-static compounds
have been commercially prepared for more than 80 years. A CPCM consists of a random
distribution of conducting filler throughout an insulating polymer; this is of interest for several
application fields. It is crucial to choose a suitable filler-polymer pair in the production of
CPCM. A good balance between filler-polymer and filler-filler interactions can lead to the
formation of a continuous network, but filler particles tend to stick together if the filler-filler
interactions dominate and aggregates rather than forming network of particle-contacting chains.
On the other hand, a good adhesion between polymer and filler results in an insulating layer
around the filler particles and prevents formation of current-conducting chains. Many conductive
polymer composites exhibit percolation characteristics(Chung, Sabo, & Pica, 1982; Flandin,
Bréchet, & Cavaillé, 2001).
The most universal conducting filler for this purpose is carbon black, because dispersion of
conducting carbon black particles in synthetic polymers is a very efficient way of providing them
with semi-conductive and antistatic properties. From the economical and technical viewpoint,
decreasing the carbon black percolation threshold as much as possible in the blend is desirable.
Consequently, the percolation theory is the most adequate for modelling conductivity of CPCM.
It has been observed that insulator-conductor transition in polymer-filler composites depend on
the aggregation(Masao Sumita, Kayaki, & Miyasaka, 1986), size, size distribution, structure and
porosity of the conducting particles(Verhelst, Wolthuis, Voet, Ehrburger, & Donnet, 1977), on
rheological properties of polymer components(Miyasaka et al., 1982b), and on processing
conditions.
The curve of conductivity versus filler concentration is S-shaped (Figure 2-18), which clearly
demonstrates a relatively narrow filler loading range during which a small increase in loading
will result in a drastic increase in conductivity. This change implies some sudden changes in the
dispersing state of conductive particles at a critical point (percolation threshold), i.e. the
coagulation of particles to form networks which facilitates the electrical conduction through the
composites. Figure 2-18 shows the conductivity of a thermoset system of multiwall-nanotube/
46

epoxy as a function of nanotube weight fraction. Consequently, insulator to conductor transition
is attributed to the critical amount of filler necessary to build up a continuous conductive
network, resulting in a tremendous change in the electrical conductivity(Du et al., 2004).
Figure 2-18. Simultaneous measurement of the nanocomposite (MWCNT/epoxy) electrical
conductivity as function of the nanotube weight fraction, performed in the liquid state prior to
curing at 0.1 s −1 and 20 °C. The rheological percolation threshold is located at 0.1 wt% and the
electrical one at 0.04 wt%(Du, et al., 2004).
Since the polymer blends containing conductive filler with continuous structures are universality
systems, the conductivity value for filler loading fractions higher than the filler percolation
threshold can be calculated with regards to the critical concentration of conductive material
(percolation threshold) pc by the following scaling law(Kirkpatrick, 1973; Stauffer & Aharony,
1994b).
Equation 2-43. σ
Equation 2-44. σσ
47

where p is the volume fraction of the conducting phase above the percolation threshold, pc is the
percolation threshold, σ is conductivity of sample at given volume fraction of p, and σc is the
conductivity of the sample at the percolation threshold. The conductivity exponent t generally
reflects the dimensionality of the system with values typically around 1.3 and 2.0 for two and
three-dimensions, respectively(Clerc, Giraud, Laugier, & Luck, 1990). Gubbels et al. found that
in a polystyrene (PS)/ polyethylene (PE)/ carbon black (CB) system, t is 2.0 when CB is
dispersed within the amorphous phase of pure PE. Its value becomes 1.3 when CB is at the
interface of a PE/PS blend with dual-phase continuity(Gubbels, Blacher, et al., 2002; Gubbels,
Jerome, et al., 2002). Al-Saleh et al.(Al-Saleh & Sundararaj, 2008) showed that selective
localization of CB at the interface of PS/PP blend is possible by introducing SBS copolymer that
can selectively localize at the interface and for which CB has the highest affinity. Addition of
5% SBS resulted in reduction of percolation threshold to less than 1%.
The classical percolation theory (Equation 2-44) applied to conductive polymer composite
systems must meet some definite conditions. First, the system should be randomly distributed,
and secondly, particles must be spherical, monodisperse and have an isotropic conductivity. If
one or all of these conditions are not fulfilled, the theoretical value will deviate far from the
realistic one. For example, Carmona et al. (Carmona, Prudhon, & Barreau, 1984) reported that in
short fiber-epoxy resin composites, the exponent t = 3.0 ± 0.6 when the length of fibers ranges
between 1.15 mm and 2.85 mm.
A simplified view of percolation transition is depicted in Figure 2-19. At low filler loadings the
conductive particles act like conductive islands in a sea of electrically insulating resin. Since
electrons moving through the composite still encounter the insulating polymer, little or no
change in conductivity is observed. Addition of further conductive material leads to the
conductive particles being more crowded and more likely to come into contact with each other.
Finally, at the percolation threshold, the first complete cluster from the bottom to the top of the
sample is generated.
48

Figure 2-19. Approaching the percolation threshold in a binary blend
Thereby, continuous chains or networks of conductive material through sample are formed. An
electrical charge can now pass through the conductive composite or blend via this network
without encountering the low conductivity of the polymer resin. Introducing further conductive
material loading beyond the percolation threshold does not greatly increase the conductivity of
this material(Bhattacharya, 1986).
2.3.1.1.1 Parameters Influencing the Percolation Threshold of Conductive
Fillers
There are several ways to decrease the percolation threshold of conductive filler concentration in
polymeric matrices, which are mainly based on the use of additives, the polymer properties, the
optimization of processing conditions, and the size, distribution and porosity of the
filler(Smuckler & Finnerty, 1974; Verhelst, Wolthuis, Voet, Ehrburger, & Donnet, 1977).
Therefore, varying these factors can cause strong deviations from the percolation theory. Several
important factors influencing the percolation threshold in conductive polymer composites are as
follows:
• Type and Shape of Conductive fillers
The shape of the particle plays a critical role in where (at what filler volume) percolation occurs.
The more structured or elaborately shaped the particle, the more likely it is to contact a nearest
neighbor and form a continuous network. Perfectly spherical fillers, which arguably have the
least elaborate and least structured shape, can require as much as 40% loading in order to reach
49

the percolation threshold. As compared in Figures 2-20a and 2-20b, by using high aspect ratio
fillers such as conductive fibers/wires, the conductivity of the polymer composite can be
achieved at a much lower filler loading level (Figure 2-20c).
Carbon-black particles are more irregularly shaped and often have long branches reaching out
from the main body of the particle. These moderately structured fillers can require anywhere
from 5 to 35% loading to reach the percolation threshold(Sichel, 1982). In carbon black filled
nitrile rubber composite, the percolation threshold lies in between 20% and 35%, while for a
short carbon fiber filled composite it ranges from 10% to 20%(Pramanik, Khastgir, & Saha,
1992). Highly shaped fillers such as carbon or stainless-steel fibers may be present in as little as
a few percent by volume in order to achieve low resistance. Gelves et al.(Gelves, Lin,
Sundararaj, & Haber, 2006) used silver and copper nanowires with a diameter of 25 nm and
aspect ratios of up to 200 and 400, respectively. Electrical percolation thresholds of 0.5-0.75
vol.% for silver nanowires and 0.25-0.75 vol.% for copper nanowires in PS nanocomposites
were found.
Figure 2-20. Schematic illustrations of conductivity establishment for polymer composites with
(a) spherical fillers, (b) high aspect ratio fillers (fiber/wire) and (c) percolation threshold (Vc) in
a resistivity vs. filler loading plot(Wong, 2003-2008)
Lin et al.(Lin, Gelves, Haber, Potschke, & Sundararaj, 2008) reported an electrical percolation
threshold between 1 and 2 vol.% for polystyrene/copper nanowire nanocomposites. The
nanowires have a diameter of around 25 nm and an average length of 1.29±0.83 μm. A very low
percolation threshold of high aspect ratio fillers for silver nanotubes is shown in Figure 2-21.
50

Figure 2-21. Transmission electron micrograph of silver nanowires(Wong, 2003-2008)
• Polymer Polarity
Some studies(Miyasaka et al., 1982a) revealed that the polymer polarity may be related to the
percolation threshold; that is the higher the polarity of a given polymer, the larger the percolation
threshold is. As well, a correlation between the surface tension of the polymer and the
percolation threshold was found(Johnson, Kendall, & Roberts, 1971): the larger the surface
tension, the larger is the percolation threshold.
• Polymer Viscosity
Percolation threshold is limited by increasing the viscosity of the polymer(Sau, Chaki, &
Khastgir, 1997), since high viscosity of the polymer results in high amount of breakdown and
degradation of the structure of carbon black owing to the high shearing action experienced
during mixing. Consequently, the formation of a conductive network throughout the matrix is
delayed and occurs at a higher concentration(Sau, Chaki, & Khastgir, 1998).
• Polymer Crystallization
High-density polyethylene loaded with conductive carbon black showed that the carbon black
aggregates tend to concentrate in amorphous regions. A major part of the carbon black
aggregates in a semi-crystalline system are located in amorphous regions. Lower percolation
51

threshold concentrations are obtained in semi-crystalline systems compared to amorphous
polymers due to geometrical restriction of the conductive pathway(Narkis & Vaxman, 1984).
• Polymer Blend Morphology
Another approach to reduce the percolation threshold of conductive fillers relies on the selective
localization of conductive phase particles in multiphase polymeric materials in one of the phases
or better yet at the interface of binary co-continuous blends(Gubbels et al., 1995). A percolation
threshold of 0.5 vol.% of carbon black was reported by compression molding of a mixture of
carbon black and polystyrene powders(Klason & Kubát, 1975). Although the carbon black
particles do not penetrate the polystyrene phase and remain essentially located at the interface
between polystyrene particles, this method suffers from poor mechanical properties and
reproducibility. Gubbels et al.(Gubbels, Blacher, et al., 2002; Gubbels, Jerome, et al., 2002)
succeeded in fabricating a multiphase material with low percolation of carbon black particles in
polymers by selectively localizing carbon black percolation threshold of conducting polymer
lowers owing to double percolated structure(Levon, Margolina, & Patashinsky, 2002; Sumita et
al., 1992). The percolation threshold of the poly(ethylene-co-vinyl acetate) (EVA)/high density
polyethylene (HDPE)/ carbon black (CB) blend is at a significantly lower carbon content than
the individually filled HDPE or EVA(Foulger, 1999). Similar results were obtained for
polypropylene/nylon/CB(Tchoudakov, Breuer, Narkis, & Siegmann, 1996).
The effect of the PE continuity fraction on the CB content is illustrated in Figure 2-22. Double
percolation is at the origin of this situation(Calberg et al., 1999):
1) Percolation of the polymer phases and thus of their interface.
2) Percolation of the conducting particles in one phase or at the interface.
52

Figure 2-22: Dependence of PE continuity fraction on the CB content in 5/95 and 10/90
Jerome et al.(Gubbels, Blacher, et al., 2002; Gubbels, Jerome, et al., 2002) conducted studies
with different binary blends, such as PE/PS and PS/PMMA, to investigate the influence of
parameters on the percolation threshold and extend the dual percolation concept to other binary
blends.
Jerome and co-workers found that due to thermodynamic forces governing the system and filler
aggregation, the classical theory of percolation (Equation 2-44) can not account even
qualitatively for most of the experimental data. For instance, the experimental t exponent is quite
different from the theoretical values and depends strongly on the morphology of the blend.
Hence, the theory needs to be modified. It was observed that for PE/PS blend, carbon black was
accumulated at the polymer interface as the result of a kinetic control leading to unstable
systems, but in PS/PMMA, due to the more polar polymer component (PMMA vs. PE), the
interfacial localization of CB particles was appropriately improved (Calberg, et al., 1999)(Figure
2-23).
53

PMMA
Figure 2-23. TEM micrograph of a 50/50 PS/PMMA blend filled with 2 wt% CB.
2.3.1.1.2 Percolation Threshold of Carbon-Nanotube Polymer Composites
Figure 2-24 clearly shows a large spread in the number of investigations for polymer/carbon
nanotubes in different polymer matrices. Figure 2-24 also shows the minimum observed
percolation thresholds C together with the number of publications for each polymer matrix.
This study shows that for all C > 0.2 wt% of carbon nanotubes, no more than two papers could
be exploited. This finding supports the belief that with optimized dispersion methods a
percolation threshold C ≈ 0.1 wt% might be obtainable for nearly any CNT/polymer
system(Bauhofer & Kovacs, 2009).
2.3.1.1.3 Disadvantages of Conductive Fillers, Especially Carbon Black
The great majority of commercial resins are compounded with conductive fillers to create
conductive composites. The most commonly used composite is 15–30% carbon black in polymer
to make containers, trays, bags, work surfaces, wrist straps, etc.
PS
54

Figure 2-24. Plot of the total number of publications per polymer system versus the minimum
percolation threshold achieved with the respective system(Bauhofer & Kovacs, 2009).
Figure 2-25. Resistivity of carbon black/polymer composite as a function of carbon black
volume fraction(Rosner, 2000)
55

Although polymer and carbon-black composites are versatile, relatively inexpensive, and
permanently conductive, there are several disadvantages related to the use of carbon black. The
possibility of contamination of clean-room environments by shedding of carbon particles, a
process known as sloughing, makes it unusable in many areas. Also, the level of conductivity for
a particular resin is hard to control because of the steepness of the percolation curve (Figure 2-
25). Thereby, resistivity levels ranging from 10 6 to 10 9 Ω are hard to reach, and also other
conductivity values are not completely reproducible.
2.3.1.2 Intrinsically Conductive Polymers
Applications of conductive polymers have been extended during the last years, with the emphasis
being to make better, smaller and faster electronic instruments. A few years ago, semiconductor
silicon was the only material used to fabricate almost all electronic devices. Around 16 million
bits of information can be stored by an advanced silicon chip in an area of less than 1 cm 2 ,
although a practical limit to the density of stored information in a chip exists. With further
decreasing of the size, overheating occurs and electronic components do not work properly. To
overcome this limitation, it has been suggested to use conductive polymers as a conventional
semiconductor. Moreover, polymers have the advantages of flexibility, lightweight, high
chemical inertness, corrosion resistivity, and ease of processing.
Intrinsically conductive polymers have caused a dramatic change in polymer architecture. A key
property of a conductive polymer is the presence of conjugated double bonds along the backbone
of the polymer. A regular chemical bond (C–C) occurs when two atoms share a pair of electrons.
Such sharing of electrons allows atoms to bind together and keeps molecules from flying apart.
A double bond denoted as (C=C) occurs when the atoms share an extra pair of electrons.
Conjugation occurs when there are single and double bonds alternating along the polymer chain
in a π-conjugated system of the conductive polymers(Gerard, Chaubey, & Malhotra, 2002;
Inzelt, Pineri, Schultze, & Vorotyntsev, 2000; Pron & Rannou, 2002). Consequently, electrically
conducting polymers are composed of macromolecules having fully conjugated sequences of
double bonds along the chains. It implies that in intrinsically conductive polymers, the polymer
56

chain provides the conductive path for the electrons. π-electrons are responsible for the unusual
electronic properties of these polymers, such as low ionization potential, low energy optical
transition, electron conductivity and high electron affinity.
The conductivity of a polymer can be significantly increased by doping it with
oxidative/reductive substituents or by donor/acceptor radicals. For example, Shirakawa et al.
reported a conductivity increase of 9 to 13 orders of magnitude for doped polyacetylene (PA).
Doping inorganic materials such as silicon involves substitution of atoms and is completely
different from doping organic polymers. Conductive polymers are doped with oxidizing or
reducing agents and chemicals that remove electrons from or add electrons to the polymer. A
simple explanation is that the oxidation or reduction changes the electronic structure of the
polymer so that it can conduct electricity. The degree of conductivity in the doping process is
related to many factors, including the polymeric structure, degree of doping, and type of dopant.
The “doped” form of polyacetylene had a conductivity of 10 5 Siemens per meter, which was
higher than that of any previously known polymer. As a comparison, silver and copper have
conductivity of 10 8 S m –1 .
One of the most important domains still at the foreground of research activity in electrochemistry
is preparation, characterization and application of electrochemically active, electronically
conducting polymeric systems. From a practical standpoint, the first generation of intrinsically
conductive polymers did not achieve great commercial success. These polymers tended to be
insoluble, unprocessable, and extremely sensitive to environmental conditions.
The genesis of the field can be traced back to the mid 1970’s when polyacetylene was reportedly
prepared by accident by Shirakawa(Ito, Shirakawa, & Ikeda, 1996) as the first polymer capable
of conducting electricity. Polythiazyl (SN)x discovered in 1975, was one of the pioneering
conjugated polymers which showed conductivity around 0.29 K(Greene, Street, & Suter, 1975).
Shirakawa et al.(Shirakawa, Louis, MacDiarmid, Chiang, & Heeger, 1977) in 1977 established
the idea of using polymers for electrical conducting goals when the iodine-doped transpolyacetylene,
(CH)x, exhibited a conductivity of 10 3 S/cm. After the appearance of the first
organic conductive materials, other conducting polymers were synthesized, such as polyaniline
(PANI), polypyrrole (PPY), poly p-phenylene (PPP), polythiophene (PT), and polyfuran (PFU),
which are shown in Figure 2-26. Researchers started to work in this area and could tune the
57

conductivity of these polymers for various applications from the insulating regime to the
superconducting regime, using different methods of chemical modification, and especially
doping.
During the last decade, conductive polymers have found applications in a wide variety of fields
of technology and science, such as biosensors for coupling of electron transfer, antistatic coating,
development of individual electronic devices, gas sensors, electrochromic windows, preparation
of pH or reference electrodes, corrosion protection, and whole integrated circuits.
At present, conductive polymers are extensively used in novel technologies involving
electrochromic, electroluminescent and optoelectric devices such as light emitting diodes (LEDs)
or flat-panel light emitting displays, electronically conductive fibers and wires, solar cells, solid
capacitors, and Schottky barrier diodes(Galal, Atta, & Mark Jr, 1998; Kumar & Sharma, 1998).
Figure 2-26. The structure of a number of intrinsically conducting polymers (ICP).
58

2.3.1.2.1 Polyaniline
One of the most technologically promising electrically conductive polymers is polyaniline, due
to its good environmental stability(Huang, D., & MacDiarmid, 1986), ease of synthesis, versatile
processability and stable conductivity in relation with other conductive polymers. Investigations
by Macdiarmid et al.(Macdiarmid et al., 1985) in the mid-1980’s resulted in the discovery of
electrical conductivity for the emeraldine salt of polyaniline, which led to an explosion of
interest in this fascinating polymer. Polyaniline is known to be crystalline, and a highly
conducting state is accomplished by simple protonation of the imine nitrogen atoms in the
emeraldine base backbone. Only one form of polyaniline, called the emeraldine salt, is
electrically conducting. The main disadvantage of all intrinsically conductive polymers,
including PANI, is its limited melt-processability. Addition of zinc compound to doped-PANI
made some complex allowed a method to tailor the processability of sulfonic acid-doped
PANI(Hartikainen et al., 2001; Ruokolainen et al., 2000). It has been observed that for polymer
blends containing PANI, a range of conductivity between 10 -10 S cm -1 to 10 -1 S cm -1 (melt
processing) and 10 -10 S cm -1 to 10 S cm -1 (solution processing) can be achieved(Panipol, 2000).
There are several dopants belonging to the sulfonic acid group for PANI, such as
dinonylnaphthalenedisulfonic acid (DNNDSA). Among the sulfonic acids, camphor-10-sulfonic
acid (CSA) is much used for this purpose due to high solubility of PANI-CSA in m-cresol(Vikki
et al., 1996). A soluble PANI-dodecylbenzenesulfonic acid (DBSA) dispersion was synthesized
by Kababya et al.,(Kababya, Appel, Haba, Titelman, & Schmidt, 1999) showing that DBSA is
molecularly miscible with PANI. PANI-DBSA dispersions have been recently used to prepare
blends with other commodity plastics(Haba, et al., 2000).
2.3.1.2.1.1 Binary PANI/Polymer or Ternary PANI/Polymer/Polymer Blends
Co-continuous polymer blends of PANI and a variety of classical polymers have been
extensively studied to improve mechanical properties of these materials and also to decrease the
percolation threshold of the conductive polymers. Polystyrene(PS)(Jousseaume, Morsli, &
Bonnet, 2002; Jousseaume, Morsli, Bonnet, Tesson, & Lefrant, 1998; Won-Jung et al., 2004;
59

Woo Jin, Won Ho, & Yun Heum, 2003), cellulose acetate(Niziol & Laska, 1999; Planes et al.,
1998; Pron, Zagorska, Nicolau, Genoud, & Nechtschein, 1997; Wolter, Banka, Genoud, Pron, &
Nechtschein, 1997), polymethyl methacrylate (PMMA)(Angappane, Rajeev Kini, Natarajan,
Rangarajan, & Wessling, 2002; Morgan, Foot, & Brooks, 2001), polyethylene (PE),
polypropylene (PP), polyamide (PA)(Domenech, Bortoluzzi, Soldi, & Franco, 2002; Jing, Yang,
Zheng, & Lan, 1999; Qinghua, Xianhong, Yanhou, Dajun, & Xiabin, 2002; Zhang, Jin, Wang, &
Jing, 2001), polyimide (PI)(Liangcai et al., 2001; Moon Gyu & Seung Soon, 2001; Su et al.,
1997; Watcharaphalakorn, Ruangchuay, Chotpattananont, Sirivat, & Schwank, 2005), polyvinyl
chloride (PVC)(Chipara et al., 1998; Laska, Zak, & Pron, 1997), polyurethane (PU)(Rodrigues &
Akcelrud, 2003; Rodrigues, Lisboa-Filho, Mangrich, & Akcelrud, 2005; Yoshikawa, Hino, &
Kuramoto, 2006), and a wide variety of thermoplastic elastomers(Dong-Uk, Kitae, Young Chul,
Yun Heum, & Jun Young, 2001; Goh, Chan, & Ong, 1998; Ong, Goh, & Chan, 1997) were used,
among others. Conductive polymer blends with good mechanical properties and conductivity
varying between 10 −11 and 300 S/cm have been easily fabricated. There are some other reasons
for blending PANI with general polymers, such as to increase the stability of conduction. It has
been observed that for polymer blends containing PANI, a range of conductivity between 10 -10 to
10 -1 S/cm (melt processing) and 10 S/cm (solution processing) can be achieved(Panipol, 2000).
Different techniques, including melt-blending, solution blending, in-situ polymerization, and
dispersion mixing have been used to prepare binary and ternary PANI/polymer (A)/polymer (B).
Each technique has some advantages and some disadvantages. For example, in PS/PANI meltblend,
the percolation threshold of PANI is much higher than that in a PS/PANI blend prepared
by dispersion mixing, although it is clear that a binary blend composed of PANI and PS prepared
by melt-blending has better mechanical properties than a PS/PANI blend made by dispersion
mixing(Segal, et al., 2001). Moreover, Woo et al.(Woo Jin, et al., 2003) reported that PS/PANI
blends prepared by an in-situ polymerization technique have a higher conductivity than by the
other techniques, such as solution blending. A conductivity value of 0.1 S/cm for 20 wt% of
PANI is observed for a PS/PANI blend prepared by in-situ polymerization, due to a better
morphology compared with the other methods, such as solution blending. Zilberman et
al.(Zilberman, et al., 2000d) reported that for PS/PANI and LLDPE/PANI, the percolation
threshold is about 30 wt% PANI. For a binary blend composed of CoPA/PANI, a value of 20
wt% PANI is obtained. They also observed a high-quality conducting PANI network in a
60

CoPA/LLDPE/PANI blend and a poor-quality PANI network in a (PS+DOP)/LLDPE/PANI
blend. They concluded and reported formation of a host matrix containing two co-continuous
immiscible thermoplastic polymer phases and developed a model predicting that doped PANI
can generate a network at the interface of CoPA/LLDPE blend and disperse in a
(PS+DOP)/LLDPE blend due to their solubility properties. Zilberman et al.(Zilberman et al.,
1997) measured the stability of conductivity of 80%PS/20%PANI-DBSA blend at 80ºC in a
vacuum and at 50ºC in 60% RH based on exposure time. It was noted that after four hours, the
conductivity decreased from 0.25 S/cm to about 0.1 S/cm in both states. Fraysse et al.(Fraysse,
Planes, & Dufresne, 2000) could decrease the percolation threshold of PANI sharply in a
PANI/PMMA blend prepared by melt dispersion to 0.5 wt% of PANI. TEM micrographs showed
that connectivity of the PANI-CSA network in PANI-CSA/PMMA blend decreases rapidly for a
volume fraction less than 0.5% (Figures 2-27a and 2-27b)(Reghu et al., 1994; C. Y. Yang, Y.
Cao, Paul Smith, & A. J. Heeger, 1993a).
A typical scenario for a percolating medium with “links” (PANI-CSA fibrils), “nodes” ((crossing
points of the links), and “blobs” (dense, multiply connected regions)(Stauffer & Aharony,
1994b)) is illustrated in these figures. Numerous links with diameters of about 100-500 Å are
clearly exhibited in the sample containing 0.5% PANI-CSA (Figure 2-27a), while rather few
links between the nodes and blobs are observed in sample containing 0.25% PANI-CSA (Figure
2-27b). They showed that the relationship between conductivity and volume fraction of PANI-
CSA was critically dependent upon the nature of mass distribution among the links, nodes, and
blobs in the samples. They found various parameters involved in this distribution including the
molecular weight of the polymers, the viscosity of the polyblend solution, the solvent, and the
drying temperature.
61

Figure 2-27. Transmission electron micrographs of extracted PANI-CSA/PMMA polyblend
films containing (a) Vol% = 0.5% , and (b) Vol% = 0.25% PANI-CSA(Yang, et al., 1993a).
Figure 2-28a depicts the percolation threshold of PANI-CSA in PANI-CSA/PMMA. The scaling
law of the percolation theory was applied for linear lines represented in Figure 2-28b to identify
the percolation threshold of such a material.
a)
a)
Figure 2-28. a) Conductivity vs volume fraction(f) of PANI-CSA at T=300K and T=10K, b)
conductivity vs (f-fc) where percolation threshold (fc) is 0.3%. The solid lines through the points
correspond to t=1.99 at T=10K and t=1.33 at T=300K(Yang, et al., 1993a).
b)
b)
62

Narkis and co-workers extended and developed previous works about melt-blending PANI and
other general polymers in order to decrease the percolation threshold and increase mechanical
properties(Haba, et al., 2000; Narkis, et al., 2000a; Segal, et al., 2001). They compared
dispersion mixing (Figure 2-29a) and melt-blending process (Figure 2-29b)(Narkis, et al., 2000a)
and found a percolation threshold of less than 1% for the former case and approximately 20% for
latter case.
a) b)
Figure 2-29: Conductivity vs. concentration of PANI/polymer prepared via a) dispersion
mixing, b) melt-blending(Narkis, et al., 2000a)
Levon et al.(Levon, Margolina, & Patashinsky, 1993) were the first who suggested utilization of
three components instead of two. They introduced the term “double percolated” in a ternary
polymer blend consisting of a conductive polymer. Double percolated system was defined as the
connectivity in a hierarchical basis of a connected path within a connected path, the last of which
is conducting. Narkis et al.(Narkis, et al., 2000a) reported a ternary blend comprised of
(PS/DOP)/CoPA/PANI and showed that CoPA situates at the interface. They showed that PANI
preferentially locates at the CoPA phase demonstrating a double percolation phenomenon in
which PANI is encapsulated by CoPA. The conductivity of the blend as a function of
concentration of CoPA was studied (Figure 2-30).
63

Solvent casting followed by a thermal imidization process was utilized to prepare PANI-
DBSA/PI blend. (Han & Im, 1999) It was observed that at a low amount of PANI (about 5 wt%),
a good conductivity can be achieved and even has a higher conductivity than pure doped PANI
as it is represented in Figure 2-31.
Two recent studies were performed on binary PANI/PVDF blends(Malmonge et al., 2006)
(Privalko et al., 2005). The conductivity curves are depicted in Figures 2-32a and 2-32b,
representing a percolation threshold of 15 wt% and 3.5 wt%, respectively.
Figure 2-30. Electrical conductivity vs. CoPA content for (PS/DOP)/CoPA/PANI-pTSA blends
of various PANI-pTSA contents(Narkis, et al., 2000a)
64

Figure 2-31: Conductivity of PVDF/PANI and pure doped PANI as a function of PANI
content(Wt%)
Using the definition of the percolation threshold, it appears that co-continuous structures with
various levels of continuity can be formed at various compositions above the percolation
threshold. There are several methods to evaluate the percentage of continuity, among which is
the relatively simple gravimetric analysis.
a) b)
Figure 2-32: Conductivity of PANI/PVDF binary blend based on PANI concentration a)
(Malmonge, et al., 2006) and b) (Privalko, et al., 2005)
65

Co-continuous polymer blends are an interesting and challenging research topic that offer
promising opportunities for improving desirable properties. These structures have potential
applications in a wide range of fields(Potschke & Paul, 2003), ranging from improvement of
physical and mechanical properties(Harm Veenstra et al., 2000) to electrical
conductivity(Geuskens, Gielens, Geshef, & Deltour, 1987; Man & Shaw, 2006; Soares et al.,
1995) or selective permeability.
2.4 Self-assembly Processes in Material Science
Self-assembly in the classic sense can be defined as the spontaneous and reversible organization
of molecular units into ordered structures by non-covalent interactions. Or more simply, self-
assembly is defined as the science of things that put themselves together. Hosokawa et
al.(Hosokawa, Shimoyama, & Miura, 1996) defined self-assembly as a system in which the
following conditions should be met: 1) some bonding forces should be generated; 2) bonding
should be selected, 3) moving the parts randomly so that they come together by chance.
Campbell et al.(Campbell, Freidinger, Hastings, & Querns, 2002) gave a simpler definition for it
which “refers to aggregating of particles into an organized structure without external assistance.”
The most straightforward definition was offered by Reif et al.(Reif, Sahu, & Yin, 2006) as a
“process in which small objects autonomously associate with each other to form larger
complexes”. Finally the most comprehensive definition was presented by Pelesko as
“spontaneous formation of organized structures through a stochastic process that involves preexisting
components, is reversible, and can be controlled by proper design of the components,
the environment, and the driving force”.
2.4.1 Layer-by-Layer Self-Assembly Technique
Polymer thin films have a great importance in a wide range of technological applications, such
as devices with tailored mechanical, physical, chemical, optical, electrical properties or
functional coatings. To achieve a well-controlled supramolecular architecture in fabricating
66

organic thin films, a novel and straightforward method was developed in the last decade(G.
Decher & Hong, 1991a; Nicolau, 1985). It was observed(G. Decher & Hong, 1991a) that
polyelectrolytes with opposite charges could be assembled in an alternating layer by layer
method onto a substrate to generate a sequential multilayer organic thin film(G. Decher, Hong, &
Schmitt, 1992b). The potential utility of this method is to obtain thin, uniform electrically
conductive films on a variety of substrates, as was shown by Rubner and his group(Cheung, Fou,
Ferreira, & Rubner, 1993; Cheung, Fou, & Rubner, 1994; Cheung, Stockton, & Rubner, 1997). It
was found that two consecutive layers are strongly interpenetrated, although some researchers
believe that the layers are locally interpenetrated and diffusion from one polyanion layer to the
neighboring polyanion layer can not occur,(Farhat, Yassin, Dubas, & Schlenoff, 1999) but some
other authors mention that a network of two co-continuous layers is generated(Cheung, et al.,
1997).
In such a film-forming technique, layer-by-layer deposition of different polyelectrolytes with
opposite charges is involved to fabricate molecularly controlled ultrathin multilayer films(Tse et
al., 2007). Electrostatic interactions are the cause of the adsorption of layers. Kinetic trapping of
charge species from solution on the surface can be described as film assembly( Decher, 1997).
Repetitive deposition steps provide a precise control over the total thickness of the layers in the
range from a few angstroms up to the micrometer range. The LbL technique, unlike other film
fabrication processes such as the Langmuir-Blodgett (LB)(Blodgett, 1934; Zasadzinski,
Viswanathan, Madsen, Garnaes, & Schwartz, 1994) technique, which requires particular
equipment, has no limitation on the type of substrate and charge-bearing species. Consequently,
LbL is a simple method and considerably more versatile than other techniques.
Decher et al.(G. Decher & Hong, 1991b) developed the LbL deposition method since it has been
extensively used to construct thin films of well-defined thicknesses for a wide variety of
applications that can be divided into two categories(Decher & Schlenoff, 2003): 1) tailoring
surface interactions, such as corrosion protection(Farhat & Schlenoff, 2002), sensors(Decher,
Hong, & Schmitt, 1992a), antibacterial(Rudra, Dave, & Haynie, 2006), antistatic coating for
electrophoresis(Graul & Schlenoff, 1999), and biosensing(Yipeng Sun, 1996), and 2) fabrication
of surface base devices such as electro-optic materials(Fou, Onitsuka, Ferreira, & Rubner, 1996),
devices including nanoreactors(Joly et al., 2000), electrochromic windows(DeLongchamp &
Hammond, 2001), or even in the area of surface patterning for the deposition of quantum dots
67

with a polyelectrolyte(Jaffar et al., 2004). In the LbL approach, the adsorption process involves
consecutive and alternate deposition of polyelectrolytes including polyanions and polycations
driven by electrostatic forces, or other interactions such as hydrogen bonding. Although
oppositely charged polyelectrolyte side groups represent a quite powerful attraction, some other
types of intermolecular interactions between polyelectrolytes such as van –der Waals, dipole-
dipole, and ion-dipole, involve chains of polyelectrolyte and molecules of solvent.
Polyelectrolytes or polysalts are similar to both high molecular weight compounds (polymers)
and electrolytes (salts). This means that due to electrolyte groups, they are electrically
conductive in aqueous solutions and because of macromolecules, their solutions are viscous. In
the LbL technique, each step involves dipping of the substrate into an oppositely-charged
solution followed by a rinsing step with water. To maintain film growth until the desired
thickness or number of bilayers is achieved, alternate dipping is repeated. A stepwise increase of
multilayer thickness or film mass with each deposition step is expected. Thickness of each
monolayer in this method is dictated by the polymer geometry; also supramolecular structures of
alternately charged polymers can be controlled by polymer combination, surface charges, and
solution parameters(Decher, 1997). The thickness increment after each deposition, referred to as
growth rate, shows a steady state regime of multilayer assembly with several nanometers
thickness for each layer.
Decher(Decher & Schlenoff, 2003) mentioned two significant advantages for layer-by-layer
assembly method: 1) surface properties of existing objects and devices can be controlled by
controlling surface functionality, and 2) thin-film devices can be fabricated in a very controlled
way by template-assisted assembly. Properties of devices can be engineered by controlling the
spatial arrangement of functionality in multimaterial-layered nanocomposites. There are some
more advantages for this technique in comparison with other methods such as its low-cost
preparation and its environmentally friendliness. Kotov(Kotov, Dekany, & Fendler, 1995) added
that by varying the total number of deposited layers, the total thickness of the coating can be
controlled with a precision of a few nanometers. Thus, the technique is exceptionally simple,
precise, versatile and scale-up friendly.
After showing the deposition of oppositely charged polymer layers, the number of published
papers on polyelectrolyte multilayer has grown exponentially. More than 200 published papers
68

on the topic have been reported in 2000. The number of published papers per year is represented
in Figure 2-33(http://www.chem.fsu.edu/multilayers/).
Publication per year
800
700
600
500
400
300
200
100
0
LbL Multilayers Publications per Year
1990 1992 1994 1996 1998 2000 2002 2004
Figure 2-33. LbL multilayers publications per year
Polyelectrolyte is not the only component for the LbL technique as various types of
macromolecular species (charged preferred), such as nanoparticles (NPs)(Caruso, Lichtenfeld,
Giersig, & Mohwald, 1998; Kotov, et al., 1995), nanowires(R. Gunawidjaja, 2006; Tang, Kotov,
& Giersig, 2002), inorganic molecular clusters(Keller, Kim, & Mallouk, 1994), organic
dyes(Nicol, Habib-Jiwan, & Jonas, 2003), polypeptides(Lvov, Ariga, Ichinose, & Kunitake,
1995), DNA(Lvov, Decher, & Sukhorukov, 1993), and viruses(Dimitrova, 2007) can be applied
as the assembly components. Kotov called the resulting material a nanoparticle-polymer hybrid
and explained that by this method, unique properties of monolayers and nanoparticles can be
combined with the good mechanical properties of polymers. LbL deposition has practically no
limitations on the type of the substrate. Multilayer fabrication by the LbL(Krass, Papastavrou, &
Kurth, 2003) assembly technique can use various interactions other than electrostatic interaction,
including hydrogen bonding(Stockton & Rubner, 1997), donor/acceptor interactions(Shimazaki,
Mitsuishi, Ito, & Yamamoto, 1997), and metal-ion coordination(Krass, et al., 2003). The
theoretical description of this theory is quite complex because of the special interactions between
the layers. There have been several attempts to model this method(Joanny, 1999; Solis, Cruz, &
M., 1999) but they have been inaccurate and certain calculations have not been represented yet.
Year
69

2.4.1.1 Fabrication of Conductive Multilayers by the LbL Method
Conducting polymers prepared by thin film technology have become increasingly important for
electrical and optical applications. Polyaniline is one of the most promising electrically
conducting polymers due to ease of synthesis, versatile processability, stable electrical
conductivity, and relatively low cost. PANI is used in a wide variety of fields such as chemical
sensors(Dan, Yadong, Zhiming, Xiangdong, & Yanrong, 2000; Mathur, Misra, Singh, &
Tripathi, 2002), solar cells, light-weight batteries, light emitting diodes, polymer actuators, and
corrosion protection agents. Shirakawa et al.(Ito, Shirakawa, & Ikeda, 1974) found that by
“doping” polyacetylene, its conductivity is increased by many orders of magnitude. There are
two kinds of dopants: electron acceptors (p-type dopants) such as iodine, and electron donors (n-
type doping). The use of various concentrations of dopants can vary electrical properties by as
much as 18 orders of magnitude, and so conductivity can be controlled in a wide range. The
conductivity of the emeraldine base (EB), which is the best representation of polyaniline at
ambient temperature, is about 10 -10 S/cm, while by doping it with different acids and adding sites
of (-N=) to EB, emeraldine salt can be formed with a conductivity of about 5 S/cm at the same
temperature(Epstein et al., 1986; Ginder, Richter, MacDiarmid, & Epstein, 1987). Ferreira et
al.(Ferreira, Cheung, & Rubner, 1994) found that both aqueous and polar organic solutions of
positively charged protonated forms of polyaniline and polypyrrole can be prepared, although
solubility of PANI in an organic medium such as dimethylacetamide(DMA) is much higher than
that in water. Nevertheless, for LbL self assembly, preferred media are aqueous solutions.
The preferential affinity of certain ions for others is an important factor in film growth. In low
molar mass compounds, chloride or bromide bind much stronger to the ammonium group than
sulfonate does and sodium binds stronger to the sulfonate group than to chloride and bromide.
For example, film growth forms smoothly using sodium poly(styrene sulfonate) and poly(N,N,Ntrimethyl-2-methacryloylethyl
ammonium) bromide or poly(diallyldimethyl-ammonium)
chloride(Bertrand, 2000).
Rubner’s group at MIT investigated both electrostatic and hydrogen-bonded LbL assembly using
fully and non-fully ionized polyelectrolyte in solution. They found that the structure of
polyelectrolyte multilayer depends on the degree of ionization of the polyelectrolyte. They
70

succeeded in creating thin films with unique and quite useful properties by using weak
polyelectrolytes and tuning charge densities by simple changes in solution pH. It was noted by
Cheung and Rubner(Cheung, et al., 1997) that under different pH environments, a weak
polyelectrolyte becomes less ionized and so a pH-sensitive trigger can be used to release the
charged molecules.
2.4.1.1.1 Charged Polymer Adsorption
In this section, the physical aspects of polymer adsorption are focused. From an architectural
point of view, polymers can be divided into linear homopolymers, branched homopolymers,
disordered linear heteropolymers, charged polymers or polyelectrolytes, and heteropolymers
composed of a backbone and side-chains of different chemical nature (Figure 2-34).
a)
b)
c)
d)
Figure 2-34. Schematic view of different polymer types: (a) linear homopolymers that are
the main subject of this review, (b) branched polymers, (c) charged polymers or polyelectrolytes
(PEs), and (d) a disordered copolymer with no specific order of the different monomers.
71

Shapes and sizes of polymer molecules are controlled by the nature of the covalent bonds that
bind the atoms of the molecule together. Generally, polymer molecules in a particular kind of
solution or in a melt state have a coil shape which is called a random coil. Solvent affinity also
affects the conformation of polymers associated with polymer architecture. Therefore, a random
conformation is usually found for an uncharged linear polymer chain in solution. In the case of
polyelectrolytes, charge also has a significant influence, as charged linear polyelectrolyte chains
repel each other due to coulomb repulsion, which causes the chain to adopt a more expanded
rigid-rod-like conformation. Persistence length is defined as a characterization factor for stiffness
on macromolecule chains. It is the length over which the normalized bond vectors at different
locations on the chain are correlated. It can also be described as an estimate for the typical radius
of curvature of chains.
a)
b)
c)
d)
Figure 2-35. Schematic view of the four scaling ranges in the Gaussian-persistent regime(Netz
& Andelman, 2003)
72

Addition of salt to the solution of polyelectrolyte will cause screening of the charges and
collapse of the rod polyelectrolyte chain occurs, favoring a more conventional conformation,
essentially identical to a neutral chain in good solvent. Another way to control the charge of each
monomer for weak polyelectrolytes, including polyacids and polybases, is adjusting the pH of
the solution. Consequently, it is concluded that due to chain stiffness and electrostatic repulsion
between monomers in a polyelectrolyte, the effective persistence length is increased compared to
uncharged polymers. Figure 2-35 shows a schematic view of the four scaling ranges in a
Gaussian-persistent regime for a polyelectrolyte structure.
As seen, on scales as small as a Gaussian blob of size Rel, each blob contains a segment of length
Lel. Within these blobs, electrostatic interactions are not so effective. Chain stiffening is caused
because of electrostatic repulsions on larger length scales. On a scale larger than Rel, Gaussian
(electrostatic) blobs are aligned linearly due to electrostatic repulsion.
2.4.1.1.2 Polymer Adsorption on the Surface
The choice of substrates has so far been dominated by their convenience for particular analytical
methods. Glass and quartz, because of their transparency, are the most suitable materials for
utilization of optical microscopy and UV-spectroscopy. Silicon wafer enables X-ray reflectivity
studies, ellipsometric studies, and IR-spectroscopic investigations. Gold-coated surfaces are well
suited for electrochemical studies, surface plasmon resonance (SPR) investigations, and studies
by quartz crystal microbalance (QCM). If the interaction between the solvent and the surface is
less favorable than that between the polymer and the surface, the polymer will adsorb from the
solution onto the substrate surface. Charged polystyrene sulfonate (PSS), which is soluble in
water, has the ability to adsorb onto various hydrophobic surfaces because of the equilibrium
adsorption, where the concentration of monomers of the polymer in the bulk solution is less than
their concentration close to the substrate surface(Klitzing et al., 1999). In the opposite case, when
the interaction between solvent and surface is more favorable than the interaction between
monomers of polymer and surface, the entropy of mixing will cause a depletion of
73

macromolecules from the substrate. Depletion of polystyrene, soluble in toluene, from a mica
substrate is an example of this case(Luckham & Klein, 1985).
The type of surface and its internal structure significantly affects the adsorption process. The
potential of the surface can be either short-range, influencing on monomers which are in direct
contact with or close to the surface, or it can be long-range, affecting on farther monomers
through van der Waals or electrostatic interactions. Various possible substrates include: 1) a flat
and homogeneous substrate (Figure 2-36a), 2) rough and corrugated surfaces (Figure 2-36b), 3) a
curved surface or spherical substrate in which curvature effects can be neglected if the radius of
curvature is much larger than the polyelectrolyte size (Figure 2-36c), 4) some other particular
surfaces, including chemically non-homogeneous substrates, interfaces of two immiscible
liquids, or surfaces with internal degrees of freedom like surfactant monolayers (Figure 2-36d).
Figure 2-36. Different possibilities of substrates:
(a) the prototype, a flat, homogeneous substrate;
(b) a corrugated, rough substrate. Note that
experimentally, every substrate exhibits a certain
degree of roughness on some length scale; (c) a
spherical adsorption substrate, such as a colloidal
particle. If the colloidal radius is much larger than
the polymer size, curvature effects (which means
the deviation from the planar geometry) can be
neglected; (d) a flat but chemically heterogeneous
substrate;
Adsorption of polymers on the substrate can be performed through various mechanisms such as
grafting of an end-functionalized polymer via a chemical or physical bond. One of the most
influencing parameters on adsorption characteristics is polymer topology. Adsorption of star
polymers and ring polymers, investigated by some research groups, have been found to be
a)
b)
c)
d)
74

different from that of linear homopolymers(Stratouras & Kosmas, 1992; Van Lent, Scheutjens,
& Cosgrove, 1987). As mentioned before, due to the increase of polymer concentration close to
the surface, polymer chains adsorb to the substrate. In an adsorption process of a linear
homopolymer, all monomers have the same interaction with the substrate, but some
distinguished sections of adsorption are still observed (Figure 2-37). An adsorbed single polymer
chain can be divided into three sections, consisting of trains, which are monomers bound to the
surface, loops, which are a number of unattached monomers between trains, and tails, which are
end sections that can form dangling sections.
Figure 2-37. The ‘tail’, ‘train’ and ‘loop’ sections of the adsorbing chain
2.4.1.1.3 Adsorption of Polyelectrolytes on the Surface of a Substrate
Generally, there are two factors determining adsorption behavior for a neutral polymer chain.
The first factor is reduction of conformational states of the polymer at the substrate. The second
one is substrate-monomers interactions(Israelachvili, 1992) that depend on the solvent, the
chemical nature of the monomer, and the substrate material. Attractive or repulsive van der
Waals interactions exist between the substrate and polymer monomers. It is observed that for
strongly attractive potentials or low temperatures, the adsorption between polymer and surface is
very strong and the thickness of the adsorbed layer is small (Figure 2-38a). Conversely, at high
temperatures or if the attractive potentials are weak, a weakly adsorbed layer is anticipated while
a diffuse layer thickness, much larger than in the previous case, is observed (Figure 2-38b).
75

a)
b)
Figure 2-38. Schematic drawing of single-chain adsorption. (a) In the limit of strong coupling,
and (b) in the case of weak coupling
Therefore, adsorption behavior of a neutral polymer chain is a result of a competition between
the attractive potential and entropic repulsion. The former tries to bind the monomers to the
substrate surface and the latter tries to maximize entropy.
In the case of the adsorption of a single semi-flexible polyelectrolyte on an oppositely charged
substrate, the adsorption process is a result of a balance between electrostatic repulsion of
charged monomers, which leads to chain stiffening, and electrostatic attraction between substrate
surface and polymer chain. It is much more complicated than the corresponding adsorption of
neutral polymers because the driving force in this case is electrostatic potential. Due to electrical
charges of polyelectrolytes, flexibility of the polyelectrolyte chains becomes an important factor
in the adsorption of polyelectrolyte to the surface. In order to measure flexibility of a
polyelectrolyte, the persistence length is defined. Flexible chains have low persistence lengths
while inflexible or rigid ones have high persistence lengths. Light scattering or viscosity
measurements are performed to measure persistence length of polymers in dilute solutions.
For charged polymers (polyelectrolytes) in the bulk, the effective persistence length is increased
due to electrostatic repulsion between monomers. Polyelectolyte conformation is governed by
intra-chain Coulombic repulsion between charged monomers. Expansion of chain structure due
to increase of the persistence length is represented in Figures 39a, 39b, and 39c.
In comparison with neutral polymers, this electrostatic repulsion results in a more extended and
swollen conformation (Figure 2-39c).
76

Figure 2-39. Macromolecule with a) small persistence length, b) medium persistence length,
and c) large persistence length
In the adsorption of a single polyelectrolyte chain on the surface, confinement in the adsorbed
layer causes an entropic repulsion. If the effective persistence length of the polymer is much
larger than the thickness of the adsorbed layer, an extended chain on the surface is observed
(Figure 2-40a). In contrast, a random coil with many loops within the adsorbed layer is generated
if the effective persistence length is smaller than layer thickness.
Figure 2-40. Schematic picture of a) the adsorbed
polymer layer when the effective persistence length is
large, (b) Adsorbed layer for the case when the
persistence length is small. In this case the polymer
forms a random coil with many loops
a)
c)
Consequently, electrostatic interactions play a major role in the adsorption of polyelectrolytes.
Other important parameters affecting electrostatic interactions, besides the charged monomers,
are the surface charge density, or surface potential in the case of conducting surfaces, the
concentration of salt in the solution, or ionic strength of low molecular weight electrolyte, and
the pH of the solution.
a)
b)
b)
77

2.4.1.1.4 Various Mechanisms of Film Growth
Understanding how the multilayers are structured and what governs their assembly is of great
importance for tuning their physico-chemical properties.
Two main modes of growth of polyelectrolyte multilayer films, namely linear and non-linear
(supralinear or sublinear), have been recognized. The extent of deposited polyelectrolyte per
adsorption step for consecutive steps distinguishes the two modes. Generally, strong
polyelectrolytes with high charge densities and low ionic strength result in a linear growth. In
this case, polyelectrolytes from the solution interact mostly with constituents of the film surface
and less with more deeply buried molecules. In linear growth, combination of different
techniques such as ellipsometry, SPM, X-ray and neutron reflectivity led researchers to establish
a three zone model of the multilayer films(Decher, 1997; Ladam et al., 2000). The first zone
consists of the first few polyelectrolyte layers being deposited close to the substrate surface.
Typically, the amount of adsorbed polymer per deposition cycle increases during the initial 1 to 5
deposition cycles, and then approaches a constant value(Gero Decher, Lvov, & Schmitt, 1994;
Lvov & Decher, 1994). It was observed that for poorly charged surfaces, this transition regime
can even stretch over much more than 5 deposition cycles(Kotov et al., 1997; Lutt, Fitzsimmons,
& DeQuan, 1998). The thickness per layer in this zone is smaller than in the second zone. The
second zone is the bulk film, which contains interpenetrating layers of polyelectrolytes. Due to
the proximity of the outer part of the multilayer, the local properties vary because of the solution
environment and the increase in surface roughness(Decher, 1997). The third zone is comprised
of one or a few polyelectrolyte layers close to the multilayer surface, including a diffuse double
layer of counterions to maintain overall electroneutrality(Ladam, et al., 2000).
On the contrary, the mechanism of exponential buildup processes is not clearly understood. The
first suggestion for the phenomenon was an increase in film roughness with the number of
deposited layers. Later experiments(Picart et al., 2002) suggested that non-linear (exponential)
growth results from diffusion of polyelectrolyte molecules inward and outward from a film,
while the film thickness is proportional to the amount of migrating polyion. The inward
polyelectrolyte diffusion step causes supralinear growth and is favored by a non-homogeneous
multilayer (Lavalle et al., 2002). Deposition of poly(L-lysine) (PLL) in the presence of a
78

polyanion such as hyaluronan (HA)(Picart, Ladam, et al., 2001) or poly(L-glutamic acid)
(PGA)(Lavalle, et al., 2002) revealed that PLL chains diffuse in and out of the film during
buildup. At the subsequent rinsing step, some of these free chains entrapped in the multilayer
diffuse outward from the film but lots of chains still remain inside. In the next step, these
remaining free chains were brought into contact with polyanions (HA or PGA) in solution. Due
to electrostatic interactions between them, free PLL chains diffuse out of the film, resulting in
the formation of PLL/polyanion complexes at the outer layer of the film. The amount of PLL
chains that diffuse out of the multilayer in presence of the polyanion in solution determines the
thickness of the new forming layer on top of the multilayer. The ability of the polyelectrolytes to
diffuse inward and outward of the films (mostly seen for weak polyelectrolytes) can lead to an
exponential growth system rather than a linear one. Film roughness is also increased with layer
number due to charge mismatch between the polyanions and polycations, demonstrating more
non-linear behavior(Bertrand, 2000). Hence, weak polyelectrolytes can display linear or
exponential growth, but strong ones always show linear behavior.
2.4.1.1.5 Effect of Polyelectrolyte Concentration on Multilayers
The available amount of polymer to be adsorbed is greater in solutions with higher polymer
concentration than in solutions with lower polymer concentration, which leads to greater
adsorption(Ferreira & Rubner, 1995). The main reason for greater adsorption in higher
concentration solutions is the presence of more anchor sites, including loops and tails, on the
substrate surface. This forces molecules to adopt a more compact orientation because of the
severe competition of the polymeric chains. In lower polymer concentration solutions, there is
less competition, and thus polymer chains straighten out, resulting in thinner layers and smaller
adsorption. It was shown(Cheung, et al., 1997) that the bilayer thickness of polyaniline/PSS has
an average value between 12 Å and 36 Å (Table 1), which increases by a factor of about three
when the polyanion (PSS) solution concentration increases from 10 -4 to 10 -2 M.
\
79

2.4.1.1.6 Effect of Molecular Weight on Multilayers
Similar growth mode and thickness have been observed for conventional polyelectrolytes with
various molecular weights such as poly(ethyleneimine) and poly(styrenesulfonate) (PSS).
Conventional or common polyelectrolytes refer to polyelectrolytes having been studied
extensively and available from commercial sources such as PSS, poly(acrilic acid) (PAA),
poly(diallyldimethylammonium chloride) (PDADMAC), and poly(allylamine hydrochloride)
(PAH).
Although previous research shows little effect of molecular weight on the polyelectrolyte
thickness, the length of the chains can influence the process in several ways. First, the possibility
of association of a long chain in solution to a formed layer is much higher than for a shorter one.
In other words, an adsorbed shorter chain can provide fewer binding sites than longer ones
because they are poorly suited to multilayer film assembly. Secondly, shorter chains are also
more prone to be released from the multilayer to the solution under conditions of
overcompensation.
Cheung et al.(Cheung, et al., 1997) reported similar multilayer thicknesses in a wide range of
molecular weights of poly(styrene sulfonic) acid, ranging from 5000 to 1000000.
Table 2-1. Relationship between concentration of the Polyanion and thickness of the Monolayers
in L-b-L deposition technique(Cheung, et al., 1997)
Concentration (M) Thickness per bilayer (Å)
0.0001 12
0.001 20
0.005 28
0.01 36
80

2.4.1.1.7 Effect of Salt on Multilayers
The presence of salt has a subtle effect. It screens the monomer-monomer repulsive interactions,
leading to enhancement of adsorption, because the interaction between the non-charged surface
and the chain does not have an electrostatic component.
Multilayers are fabricated using salt-containing or salt-free solution. Salt concentration has also
been taken into account as the most important factor in multilayer thickness ( Decher, 1997;
Decher & Schmitt, 1992). In salt-containing solutions, the salt ions of opposite charge
(counterions) are drawn to the charged object and form a loosely bound counterion cloud around
it. They effectively reduce or screen the charge of the object. The effective electrostatic
interaction between two charges in the presence of salt ions is governed by Debye-Huckle law.
There are two scenarios for charge balance in a system with and without salt, as shown in Figure
2-41.
a)
b)
Figure 2-41. Deposition of layer in LbL process a) in the absence of salt and b) in the
presence of salt
In the salt-free case, a positively charged polyelectrolyte is balanced by a negatively charged
one. In an alternative mechanism, for a salt-containing system, charged polyelectrolytes are
balanced by salt counterions derived from the bathing solution and construct multilayers.
Thereby, multilayers containing salt ions are thicker, less interpenetrating, and chains have a
high degree of mobility, leading to less stable structures.
Schlennof et al.( Dubas & Schlenoff, 2001a; Schlenoff & Dubas, 2001) performed sequential
adsorption of polyelectrolytes in the presence of various concentrations of salt (NaCl) from 0 to
1M. They reported that for strong polyelectrolytes, addition of further salt to the solution results
81

in swelling of the multilayer, and excess charge can be spread more efficiently into the
multilayer surface. Due to the increase in the salt concentration, the surface charge increases for
PSS/PDADMAC multilayers (Figure 2-42).
The surface charge depends on the last adsorbed layer, allowing a degree of control over surface
and interface properties. In this case, charge overcompensation is the key to multilayer
construction. The last-added polymer charge compensates the previous layer at the multilayer
surface by a factor of φ. Polymer charges within the bulk of the multilayer balance with a 1:1
stoichiometry-termed intrinsic charge compensation if φ=1 and equal φ for both positive and
negative polymers. Hence, a value of φ-1 represents an overcompensation factor. Normally, the
phenomenon of overcompensation of the surface charge occurs when polyelectrolyte chains
deposited on the multilayer surface compensate the previous layer charge while it has extra
charge to compensate the next deposited layer. In other words, overcompensation can be
described for a polyelectrolyte adsorbing on an oppositely charged surface as an overcharging
created during repeating of alternating positive and negative dipping. The overcompensation has
a maximum value at the multilayer surface and decreases towards the substrate surface. The level
of overcompensation at the multilayer/solution interface decreases in an exponential fashion
from the interface into the multilayer, representing a characteristic length of decay.
Figure 2-42. Surface charge in LbL process as a function of salt concentration(Schlenoff &
Dubas, 2001)
82

2.4.1.1.8 Overcompensation of the Multilayer/Solution Interface
The charged surface, after each new polyelectrolyte deposition, is the driving force for buildup of
multilayers in a linear growth system. In the case of exponentially growing films, at the end of
each step, on the outer surface of the multilayer (multilayer/solution interface), an extra charge
(overcompensation) appears, which is responsible for the non-linear growth of thickness.
There are two reasons for overcompensation at the interface of multilayer and solution:
overcompensation due to diffusion of polyelectrolytes in and out of the multilayer(Lavalle, et al.,
2002; Lavalle et al., 2004) and overcompensation due to addition of salt in solutions(Dubas &
Schlenoff, 2001a, 2001c; Schlenoff & Dubas, 2001).
2.4.1.1.8.1 Overcompensation due to Difussion of Polyelectrolytes inside
Multilayers
Lavalle et al.(Lavalle, et al., 2002) proposed a model based on AFM observations for
overcompensation on multilayer surfaces due to diffusion of polyelectrolytes inward and
outward throughout the multilayer film, leading to exponential growth. Later, they verified their
idea by means of confocal laser scanning microscopy (CLSM), optical waveguide lightmode
spectroscopy (OWLS), and quartz crystal microbalance (QCM)(Picart, et al., 2002). They
reported that in a PGA/PLL system, if the outer layer is PLL, two kinds of PLL chains in the film
were defined. The first population consists of chains which have formed the multilayer by being
strongly involved in interactions with PGA. The second one consists of some mobile chains
which are weakly bonded to the polyelectrolyte structure in the film construction. The latter
chains can diffuse within the film and are responsible for excess charge of multilayer
(overcompensation), which must be compensated by counterions. Dipping the multilayer in a
solution of PGA in the next step results in a strong interaction between the PGA chains and the
first population of PLL chains located at the surface, to form the outermost layer. At the same
time, mobile PLL chains in the interior of the multilayer slowly diffuse outward. When these
chains reach the surface, an extra reaction between them and PGA chains present in the solution
occurs and an extra layer of PLL/PGA is formed. These neutral complexes are weakly anchored
83

on the previously formed layer, establishing a larger structure. At this step, while most of the
PLL chains have diffused out onto the surface, free mobile PGA chains present in the solution
start to diffuse into the multilayer. This time, charge overcompensation is the result of an excess
amount of PGA chains in solution. Again, free PGA chains diffuse out, and these processes
repeat themselves alternately, resulting in a continuous film reconstruction and leading to an
increase in the thickness of the multilayers. Part of the free chains of either polyelectrolyte come
out during the rinsing step with a free aqueous solution. The increase of thickness at each step is
proportional to the mass of PGA/PLL complexes that form on the surface of the multilayer,
which are proportional to the free mobile PLL or PGA chains with weak bonds, that can easily
diffuse out of the film.
a)
Figure 2-43. Penetration of polyelectrolyte chain inward and outward of a multilayer film
Some other polyanion/polycation systems showing linear evolution in thickness increment
generate denser multilayer structures and avoid polyelectrolyte diffusion inside the film. Two
b)
f) e) d)
c)
84

important factors determining the growth regime of a system are ionic strength and pH of the
solution. Varying these parameters can change a linear growth regime for a given system to an
exponential system(Lavalle, et al., 2002).
2.4.1.1.8.2 Overcompensation due to the Addition of Salt
In the second case, salt concentration is responsible for overcompensation of the multilayer
surface. Salt concentration of the solution used for deposition has been found to be the most
important variable determining mass deposited and multilayer thickness(G. Decher, 1997). Ion
exchange phenomena in multilayers occur and displace small salt counterions by charged
polymer segments. Displacement of salt ions and polymer charge pairs compensates any excess
surface charge. Salt counterion profile visualizes overcompensation gradient as the excess
polymer charge is balanced by small salt ions(Na + or Cl - )(Klitzing & Moehwald, 1995). At lowsalt
concentrations, the surface charge is completely compensated by the polymer charge. In the
absence of salt, the excess surface charge vanishes since the electrostatic free energy diverges. At
high-salt concentration, the salt compensates some of the surface charge, resulting in a decrease
in the amount of adsorbed polymer.
This has been described in detail for two cases: a strong polyelectrolyte system
(PSS/PDADMAC)(Schlenoff & Dubas, 2001) and a weak polyacid (PAA/PDADMAC)(Dubas
& Schlenoff, 2001a), which exhibits gradual dissociation. In both cases, adsorbed amount is
determined by the surface charge due to salt concentration. In the former case, a constant surface
charge during multilayer buildup for strongly dissociated polyelectrolytes was observed. In this
mechanism, on each step the level of overcompensation is constant and proportional to a steadystate
thickness increment, and only the surface charge signs reverse. Addition of salt to the
solution increases the overcompensation charge. In the case of strong polyelectrolyte systems,
overcompensation is related to salt concentration and proportional excess charge penetration into
multilayer controls the film growth.
Addition of salt within the polyelectrolyte solution causes an ion exchange process and
polymer/polymer ion pairs are forced apart. At this step, small salt counterions displace charged
polymer segments and make polymers highly swollen because of the water of hydration which is
brought in with the salt. In this case, it is required to have a few polymer/polymer ion pairs that
85

are barely sufficient to hold the polyelectrolyte complex together. If the interaction of the
polymer/polymer ion is not strong enough, the polymer chains are sufficiently extrinsically
compensated by salt counterions, resulting in an inadequate intrinsic interaction to form and
build polyelectrolyte multilayers. When salt ions swell the polyelectrolyte at the surface, some
additional polymer adds to the surface of the multilayer due to the concentrated charges in that
area. Thus, the surface charge of the multilayer becomes overcompensated and the sign of the
surface charge reverses. Consequently, this causes an inversion of the population of surface
counterions from anions to cations or vice-versa. This step allows the oppositely charged
polymer to deposit on the surface. Figure 2-44 represents two oppositely charged
polyelectrolytes when extrinsic compensation occurs by addition of salt. It shows swelling of the
multilayer due to the added ions.
a)
b)
c)
d)
Figure 2-44. Deplacement of polyelectrolyte ions with salt counter-ions
86

As shown in Figure 2-44, swelling and overcompensation steps occur simultaneously during
construction of the multilayer, while in the case of high salt concentration, deconstruction of
polyanion/polycation is observed.
As more salt is added, the multilayer swells more and excess charge can be spread into the
multilayer surface to a point where all the polymer/polymer ion pairings are forced apart and
intrinsic compensation is no longer enough to keep the multilayer together. The critical salt
concentration for strong polyelectrolytes is much larger than for weak polyelectrolytes. It can be
said that further addition of salt can neutralize and thus dissociate weak polyelectrolytes at a
lower salt concentration(Dubas & Schlenoff, 1999). For example, the critical amount of NaCl for
PSS/PDADMAC is 3.5M, while for a weak polyacid like PAA with a molecular weight of 84500
and a pH of 11, its value is 0.3M(Dubas & Schlenoff, 2001a). Salt can easily swell weak
polyelectrolytes in the multilayer structure, like PAA/PDADMAC, and consequently give more
deposited mass on the surface while it is less stable than films constructed by strong
polyelectrolyte swollen due to salt, such as PSS/PDADMAC.
Some models(Schlenoff, Ly, & Li, 1998) have been developed for the overcompensation at the
surface resulting of polymer excess charge due to the presence of salt. It is concluded that salt
causes the swelling of the polyelectrolytes and relieves some of the steric repulsive interactions
leading to enhanced degrees of freedom. It allows a wider range for the distribution of excess
charges and a greater thickness increment per layer. Lefaux et al.(Lefaux, Zimberlin, Dobrynin,
& Mather, 2004) studied the dependence of multilayer growth rate on the solution ionic strength
by the spin assembly method. UV-Vis and AFM showed that the growth rate of polymer surface
coverage increases rapidly with salt concentration up to 0.1M, and that at a higher concentration
of salt a constant value or plateau is reached.
Another important effect of salt on multilayer thickness growth is that the addition of salt results
in an enhancement of screening and serves to decrease polymer/surface interactions, permitting
more chain loops and tails formation.
87

2.4.1.1.9 LbL Deposition on Curved Surfaces
Numerous investigations for adsorption of polymer on rough and corrugated(Blunt, Barford, &
Ball, 1989) or sinusoidal(Hone, Ji, & Pincus, 1987) substrate surfaces have been performed. The
first investigations on macroscopically curved bodies was carried out by Alexander(Alexander,
1977). Of considerable interest is the adsorption of strongly charged polymers on oppositely
charged spheres(Golestanian & Sens, 1999; Gurovitch & Sens, 1999; Wallin & Linse, 1996) and
cylinders(Kunze & Netz, 2002; Odijk, 1980). It has been found that when the radius of the
curved surface is much larger than the polymer size, this effect can be neglected. But when the
polymer size is large enough, surface curvature has a great effect on polyelectrolyte adsorption
on the substrate, whereas the electrostatic energy of the adsorbed polyelectrolyte layer depends
strongly on curvature. Very small spherical substrates need a large amount of electrostatic energy
to bend a charged polymer around them(Mateescu, Jeppesen, & Pincus, 1999; Nguyen &
Shklovskii, 2001). Curvature disfavors adsorption of long, strongly charged polyelectrolytes at
too low salt concentration, due to self-repulsion of the charged chains.
2.4.1.1.10 Effect of Roughness on the Multilayer Thickness
The adsorbed polymer layer can be described in terms of loop and tail size distributions (Figure
2-37). It has been observed that the thickness of the adsorbed layer increases with an increase in
the adsorption of loopy structures, leading to the deposition of correspondingly large amounts of
polyelectrolyte. The relative weight of loops and tails in an adsorbed layer is a measure of
entanglement of free polymers to this layer. Loops due to binding to the substrate at both ends
are more prone to entanglements than tails. On the other hand, polymer loops dangling from the
surface into the solution may entangle with free chains. Consequently, loop size distribution
governs coupling between the bulk and the interface(Adjari et al., 1994). The monomers close to
the wall mainly belong to loops, whereas the external part of the layer is mostly built up by the
tails. This fold-forming of polymer chains (loops and tails) leads to an increase in film
roughness. As the polymer concentration increases, the film surface roughness increases due to
entangling more polymer chains to loops and tails. Surface roughness is also enhanced by
88

increasing the pH value due to a decrease in electrostatic repulsion and the prevention of
molecular packing when the pH is raised(Braga, Paterno, Lima, Fonseca, & de Andrade, 2008).
It can be concluded that the increase of loops and tails is proportional to the increase of surface
roughness and adsorbed layer thickness.
2.4.1.1.11 Important Views on PANI/PSS LbL Adsorption
General aspects: Rubner et al.(Cheung, et al., 1997) reported that adsorption of polyaniline onto
a negatively charged surface occurs very rapidly. For the most concentrated solution, it took 5
min to reach complete adsorption, and in case of the least concentrated solution (around 10 -4 M)
it took approximately 25 min. It was also mentioned that even for adsorption process as long as
24 hrs, no additional net adsorption was observed. Adsorption of bilayers of PANI/PSS has a
linear rate and is independent of the number of layers deposited onto the substrate surface.
Undoped and partially doped polyaniline can be dissolved in a wide variety of solvents, but
doped polyaniline can be dissolved only in very strong solvents such as chromic acid. The poor
nature of PANI-solvent interactions favors the spontaneous adsorption of polyaniline chains onto
a variety of quite different substrates, including both hydrophilic and hydrophobic surfaces. It
also demonstrates a significant aggregation of the polyaniline chains after 1-2 weeks. The film
deposition process is very sensitive; films made by three different operators at the same
conditions revealed three completely different thickness of PANI(Braga, et al., 2008).
pH: The pH of the solution has a crucial influence on the stability of the solutions, doping and
the adsorption process. A polyaniline solution is not stable for more than a couple of hours. No
deposited layers have been observed by alternately dipping the substrate into a neutral
polyaniline solution and a poly(styrene sulfonic) acid solution. The lower pH of a polyaniline
solution has more positive charges because of the protonic acid doping, resulting in an increase
of polyaniline adsorption by a factor of around 1.5. Consequently, a higher electrostatic repulsion
is made between charges of polyelectrolytes such as PANI(Ferreira & Rubner, 1995). Polymer
chains according to this electrostatic repulsion tend to straighten out and improve molecular
packing. For this reason, a small number of PANI chains can approach the substrate surface and
the amount of polymer adsorbed decreases. On the contrary, raising the pH results in a decrease
89

in electrostatic repulsion and an enhancement of the amount of adsorbed polymer(Mattoso et al.,
2005; Raposo & Oliveira Jr, 2002). Several studies(Cheung, et al., 1997; M. K. Ram, M.
Salerno, M. Adami, P. Faraci, & C. Nicolini, 1999) did not succeed in assembling PANI/PSS
films in the presence of PANI in its neutral state.
Conductivity of PANI/PSS: some researchers have reported that with an increase of the number
of adsorbed PANI/PSS bilayers, electrical conductivity of the multilayer increases until a
saturation plateau is reached around the 13 th bilayer or the 25 th layer(Braga, et al., 2008;
Paloheimo, Laakso, Isotalo, & Stubb, 1995). Further deposition of layers showed no more
increase in the film’s electrical conductivity.
2.4.1.1.12 Characterization Techniques for LbL Deposited Layers
Commonly used methods used to characterize the internal structure of the multilayer and film
thickness could not be used in this research due to several limitations. Special techniques such as
UV-Vis, ellipsometry(Schwarz, Eichhorn, Wischerhoff, & Laschewsky, 1999), quartz crystal
microbalance (QCM)(Baba, Kaneko, & Advincula, 1999), streaming potential measurements
(SPM)(Schwarz, et al., 1999), atomic force microscopy(AFM)(Caruso, Furlong, Ariga, Ichinose,
& Kunitake, 1998), gravimetric measurement(Roy, et al., 2006), and x-ray and neutron
reflectivity(Schmitt et al., 1993) have been utilized to measure thickness or mass deposited on
the surface of each layer. Although the most commonly used method is UV-Vis, measurement of
light absorbance is impossible due to the black color of the PANI solution. .
As well, ellipsometry is one of the most appropriate techniques for showing the growth rate. As
the technique works with the reflection of emitted light, it is limited for substrates with curved
surfaces.
90

CHAPTER 3 - ORGANIZATION OF THE ARTICLES
To achieve the main objective of this work, an elegant technique to reduce the continuity
and/or electrical percolation threshold of a conductive polymer is developed by increasing the
number of continuous structures in polymer blends. Polyaniline and four other commercial
polymers with specific surface tensions, polarity and interfacial tensions are precisely selected.
The interfacial tensions between various components show a range from 1 mN/m for
PVDF/PMMA to 26.9 mN/m for PANI/HDPE.
Morphological sample characterization is performed by both SEM and FIB-AFM techniques.
The polymer blends are prepared in a melt-blending process in an internal mixer at 50 rpm for 8
min at 200°C. Selective solvent extraction of phases either assists in better detection of phases in
order to collect qualitative data, or is employed to produce porous samples. Combination of FIB
treatment and AFM technique allows us to clearly distinguish the phases based on the
topographical contrast formed by the difference in FIB etching rate of materials. In order to study
the effect of viscosity ratio on the ternary blends, both low molecular weight and high molecular
weight poly(methyl methacrylate) are selected. Fully interconnected porous materials are used as
substrates for alternate deposition of PSS and PANI.
In the first step, three components, namely high density polyethylene (HDPE), polystyrene
(PS), and poly(methyl methacrylate) (PMMA), are carefully chosen to satisfy the
thermodynamic conditions. The positive spreading coefficient of PS over PMMA predicts that
the PS phase situates at the interface of the HDPE and PMMA phases, and a complete wetting
case occurs. Experimental results confirms the prediction of the morphology by the spreading
theory as the identification of the phases can be clearly seen by the topographical heights
induced by FIB etching and subsequently quantified by AFM analysis in the topographical mode.
Addition of fourth phase (PVDF) to the ternary blend with complete wetting of
HDPE/PS/PMMA, provided that the entire set of thermodynamic spreading equations is
satisfied, results in hierarchically ordered phases of HDPE|PS|PMMA|PVDF. The Harkins
spreading theory predicts a hierarchically ordered structure of HDPE|PS|PMMA|PVDF|PANI by
addition of PANI to a quaternary HDPE|PS|PMMA|PVDF blend. Experimental results show a
very good consistency between these predictions and qualitative results.
91

After a theoretical determination of the order of all phases, controlling the composition of
phases allows us to generate novel morphologies such as onion morphology and multi-percolated
structures. SEM micrographs confirm onion morphology for a quaternary HDPE/PS/PMMA/
PVDF blend with a matrix of HDPE and PVDF droplets encapsulated by first-shell PMMA
encapsulated by second-shell PS. Changing the composition of phases, controls the evolution of
the morphology from an onion structure to a multi-percolated one. It illustrates that in a multipercolated
structure, all phases are fully interconnected and interpenetrated. In order to
quantitatively find the precise region of multi-percolation, solvent extraction followed by a
gravimetric measurement was used to obtain the percent continuity of each phase. Addition of
PANI to highly continuous quaternary blend of HDPE/PS/PMMA/PVDF demonstrates a
quadruple-percolated morphology with the order of phases as HDPE/PS/PMMA/PVDF/PANI.
As PANI situates as the inner phase, even at low concentration it is forced to spread between
other percolated phases.
In another part of the first paper, the effect of the addition of PS-co-PMMA to quinary
HDPE/PS/PMMA/PVDF/PANI blend with a quadruple-percolated structure is studied. Addition
of a compatibilizing agent of two adjacent phases (PS and PMMA) demonstrates phase reduction
of PS and PMMA in the blend. Since all phases are assembled in a hierarchically ordered
manner, reduction of the phase size of two layers forces the other assembled layers to decrease in
size. The first paper evaluates the effect of the number of the phases in a multi-percolated
structure. Ternary, quaternary, quinary, and 6-component blends containing 5% PANI are
prepared. The results indicate that increasing the number of components reduces the conductivity
percolation threshold of all phases including PANI due to geometrical restriction of the phases.
In order to spread the PANI, at least four components in the multi-percolated structure are
required. Addition of a further number of phases increases conductivity correspondent to
continuity of PANI only slightly.
In the second paper, through development of multi-percolated morphology associated with a
layer-by-layer (LbL) approach, a novel 3D porous polymeric conducting device (PPCD) is
generated. This technique allows for the percolation threshold concentration of polyaniline
conductive polymer (PANI) to be reduced to values as low as 0.19%. A fully interconnected
porous HDPE substrate with ultra low surface area is prepared by generating double- and triplepercolated
systems followed by annealing and selective solvent extraction of phases. As
92

discussed above, HDPE, PS, PMMA, and PVDF components are shown to produce a multi-
percolated structure with hierarchically ordered phases, since the interfacial tensions of the
various pairs satisfy the positive conditions of the Harkins equation sets. Ternary blends and
quaternary blends are prepared. After quiescent annealing, a substantial increase in the average
phase size is observed. Selective solvent extraction of multi-percolated structures results in
preparation of substrates with higher void volume. Hence, a fully interconnected porous HDPE
substrate of low surface area is prepared by employing double- and triple-percolated
morphology. The extra-large pores of HDPE substrates also facilitates the penetration of the
polyaniline (PANI)/poly(styrene sulfonate) (PSS) solution inward to the interconnected porous
area. Up to thirty-eight PSS and PANI layers are deposited on the internal surface of the 3dimensional
porous polymeric substrate, revealing an inter-diffused network conformation.
Microstructural information collected by SEM and AFM depicts a relatively thick
polyelectrolyte multilayer as wide as 5.5 μm for 38 layers of PSS and PANI on the surface of the
porous HDPE substrate with a void volume of 66%. The mass deposition profile of PANI/PSS as
a function of layer numbers demonstrates an unusual non-linear growth with an oscillatory
behavior. The oscillating deposition indicates that PSS and PANI polyelectrolytes diffuse in the
previously deposited layers. Like chains make contact with other like chains, and consequently,
an inter-diffused network of PANI and PSS is generated. Since salt is generally added in
procedures involving LbL deposition, the effect of the addition of 1 molar NaCl salt to the PSS
polyanion in the multilayer construction is studied.
Conductivity measurements show that the percolation threshold of PANI in porous devices is
achieved at a maximum of 8 layers. This corresponds to 0.19 wt.% PANI for a porous substrate
made of 33% HDPE/33% PS/33% PVDF and 0.28 wt.% for a porous substrate generated from
33% HDPE/33% PMMA/33% PVDF. It is found that sample conductivity increases by
increasing the number of deposited PSS/PANI layers until a conductivity saturation plateau is
reached after 32 deposited layers. Compression of porous samples can be employed as another
control parameter to achieve a wide range of conductivities in these conductive porous devices.
In the last paper, ternary HDPE/PS/PMMA blends with various compositions of components
represent a complete wetting case with the development of a thermodynamically stable PS layer
between the HDPE and PMMA. Four thermodynamically stable sub-classes of morphologies
exist for HDPE/PS/PMMA, depending on the composition of phases: a) matrix/core-shell
93

dispersed phase (60/10/30 HDPE/PS/PMMA); b) tri-continuous (40/25/25 HDPE/PS/PMMA); c)
bi-continuous/dispersed phase (50/25/25 HDPE/PS/PMMA); d) matrix/two separate dispersed
phases (20/60/20). Solvent extraction/gravimetry is used to examine the extent of continuity of
the components. Increasing the concentration of the core phase in the matrix/core-shell
morphology results in the coalescence of core phases which leads to the formation of a tri-
continuous morphology. Tri-continuous structures can be classified into two categories: type I
and type II. The bi-continuous/dispersed phase morphology consists in a co-continuous structure
for two phases with the third phase present as a dispersed phase.
A triangular composition/morphology diagram comprised of various compositions of
HDPE/PS/PMMA is prepared to better understand the composition dependence of the various
morphological states for complete wetting and their inter-conversion. This diagram demonstrates
the concentration regimes for various morphological regions. These regions are detected by a
combination of both SEM and FIB-AFM and through the use of selective extraction/gravimetric
analysis as a quantitative technique to detect inversion points. Triangular continuity plots are
introduced to show the continuity of the PS and PMMA phases respectively. It is found that in a
ternary blend, percolation threshold becomes a line due to simultaneously changing the
concentration of the two other phases in blend. Two compositional lines, or schemes, across
the 3-axes triangular diagram which traverse multiple morphological states, are selected to
study the effect of the concentration of the inner (HDPE) and middle (PS) phases on the
morphology of the middle phase and the effect of the composition of the inner phase (HDPE)
and the outer phase (PMMA) on the morphology of the middle phase (PS).
It is shown that extreme changes in the viscosity of the PMMA phase (L-PMMA and H-
PMMA) in a PS/HDPE/PMMA blend of constant composition still results in morphological
structures where the PS separates HDPE and PMMA. In both the L-PMMA case and the H-
PMMA case, the morphological classification of the bi-continuous/dispersed PMMA phase is
unchanged. It is found that the scale of the resulting morphological structures has been
significantly modified by the viscosity of the PMMA, as the system attempts to respond to the
dramatically reduced interfacial area of the H-PMMA droplets. It is also found that the
composition, interfacial tension, and viscosity ratio play an important role in determining the
phase size in ternary polymer blends.
94

CHAPTER 4 - LOW PERCOLATION THRESHOLD CONDUCTIVE
DEVICE DERIVED FROM A FIVE-COMPONENT POLYMER BLEND
4.1 Abstract
In this work we report on the preparation of a solid, 3D, low percolation threshold conductive
device prepared through the control of multiple encapsulation and multiple percolation effects in
a 5 component polymer blend system through melt processing. Conductive polyaniline (PANI) is
situated in the core of the 5 component continuous system comprised of high-density
polyethylene(HDPE), polystyrene(PS), poly(methyl methacrylate)(PMMA) and poly(vinylidene
fluoride)(PVDF) and, in this fashion, its percolation threshold can be reduced to below 5%. The
approach used here is thermodynamically controlled and is described by Harkins spreading
theory. In this work the detailed morphology and continuity diagrams of binary, ternary,
quaternary and finally quinary systems are progressively studied in order to systematically
demonstrate the concentration regimes resulting in the formation of these novel multiple-
encapsulated morphological structures. Initially, onion-type dispersed phase structures are
prepared and it is shown that through the control of the composition of the inner and outer layers
the morphology can be transformed to a hierarchical-self-assembled, multi-percolated structure.
The influence of a copolymer on selected pairs in the encapsulated structure is also examined.
The conductivity of the quinary blend system can be increased from 10 -15 S cm -1 (pure PE) to 10 -
5 S cm -1 at 5 % PANI and up to 10 -3 S cm -1 for 10% PANI. These are the highest conductivity
values ever reported for these PANI concentrations in melt processed systems.
4.2 Introduction
Over the last 30 years the polymer-polymer blending of conventional polymers has received
significant attention since these materials can result in highly synergistic property sets. Although
a vast number of studies have been devoted to the study of immiscible polymer blends and of the
95

dispersed phase-matrix, fibrillar, and co-continuous morphologies, relatively few studies have
focused on the fundamentals of three or more components with complex morphologies(Moussaif
& Jérôme, 1999; Reignier & Favis, 2000, 2003a; J. Zhang, et al., 2007).
Ternary A/B/C systems can exist in two possible states. One is known as complete wetting and
the other is partial wetting. Complete wetting can result in B and C droplets individually
dispersed in an A matrix. It can also result in the case where one B or C phase completely
engulfs the other in a matrix of A. In this latter case, the minimization of interfacial free energy
occurs when, for example, phase B is situated at the interface of, and completely wets, phases A
and C. Phase B prevents any contact between phases A and C(Figure 4-1a). In the partial wetting
state, all three phases have an interface with each other(Figure 4-1b). In such a case intact
droplets of B, for example, can be placed at an A/C interface. Recently some very novel
structures have been generated via partial wetting(Virgilio, Desjardins, et al., 2009; Virgilio,
Marc-Aurele, & Favis, 2009).
Strict thermodynamic conditions need to be met in order to have either discrete phases,
encapsulated structures or partial wetting. Torza and Mason(Torza & Mason, 1970) and then
Hobbs et al.(Hobbs, et al., 1988) employed a modified Harkins spreading theory (Equation 4-1)
to predict whether the morphology of a ternary blend is dominated by complete wetting or is
partial wetting.
λ = γ −γ
−γ
Equation 4-1. 12 23 13 12
where λ is the spreading coefficient, γ represents the interfacial tension for various polymer pairs
where the sub-indexes refer to each component. If one of the spreading coefficients such as λ12
has a positive value, a complete wetting case occurs in which phase 1 separates phases 2 and 3
(Figure 4-1a). In another case, where all spreading coefficients have negative values, the system
demonstrates a partial wetting case in which all phases have contact with each other(Figure 4-
1b). It has been shown that Harkins equation is, on the whole, a good criterion to predict the
position of phases in ternary polymer blends(Reignier & Favis, 2000, 2003b; Virgilio,
Desjardins, et al., 2009; Zhang, et al., 2007). As the number of phases increases to four however,
it is difficult to determine the morphology by such a simple model. To date, there is no equation
96

or model which predicts where the phases will situate in a multi-blend system comprising more
than three phases.
a)
λB/A > 0
λB/A < 0
Figure 4-1. Schematic representation of (a) the complete wetting case where phase B spreads
between, and fully separates, phases A and C and (b) the partial wetting case all phases are in
contact with each other
Co-continuous morphologies represent the special case where, in an A/B system, both
components are fully continuous within the blend. This type of system is referred to as a singlepercolated
structure. Recently(Zhang, et al., 2007; Zilberman, et al., 1998, 2000d), some papers
are examining the potential of double percolated structures by locating a phase with a specific
characteristic at the interface of two other continuous phases. For example, Zhang et al.(Zhang,
et al., 2007) by employing Harkins equation and controlling the composition of phases,
developed a double-percolated structure in which polystyrene was situated at the interface of
high-density polyethylene and poly(methylmethacrylate). All three phases were shown to be
fully continuous.
The most well-known utilization of a double-percolated morphology is locating intrinsically
conductive particles or polymers at the interface of binary blend of common polymers to produce
a conductive polymer composite(Anand, Palaniappan, & Sathyanarayana, 1998; A. Bhattacharya
& De, 1999; Pud, Ogurtsov, Korzhenko, & Shapoval, 2003). Classic percolation theory is
b)
97

usually utilized to explain the transition from discontinuous phase behavior to continuous phases.
In conductive materials, this transition point is defined as the critical concentration of conductive
phase required to build up a first conductive pathway and is referred to as the percolation
threshold. The percolation threshold is generally defined as the onset of long-range connectivity
in random systems. By employing a double percolated structure, the percolation threshold of the
middle phase sharply decreases as its presence is limited only to the interface. The conductivity
of a conductor-insulator blend near the percolation threshold point is calculated by the following
equation:
Equation 4-2. ( ) t
σ σ p − p
= 0
where σ is the conductivity at the percolation threshold point, σ is the conductivity of the
0
blend, p is the concentration of the conductive material, pc is the percolation threshold
concentration, and t is a dimensionless index.
The classic percolation threshold for a random dispersion of hard-core spheres in a matrix in
three dimensions is approximately 16 volume percent(Scher & Zallen, 1970). Restricting the
dispersed phase spatially, for example in interfacial tension driven structures such as a double-
percolated morphology, can significantly reduce the percolation threshold of that phase. In that
case the particles or polymer components can be precisely located at the interface leading to
dramatic reductions in the percolation threshold value(Gubbels, et al., 1995; Gubbels et al., 1994;
Zhang, et al., 2007). In other words, by controlling the morphology of the blend and restricting
the pathways of the components, the percolation threshold deviates from the standard value of
percolation threshold in a random system predicted by percolation theory. Moreover, using the
same approach, controlling the morphology of the system can also influence the behavior of
conductive systems. Confining the pathways of a conductive material in a multiphase system can
result in high conductivities at low concentration of conductive material. The dimensionless
index t in Equation 4-2 represents the effect of the morphology of the system on the
conductivity behavior. This has already been shown to be possible for conductive systems where
the conductive phase was selectively localized in the multiphase blend and consequently the
c
98

concentration threshold for the onset of electrical conductivity was significantly
decreased(Anand, et al., 1998).
In the case of conductive fillers, both the size and shape of the particles influence the percolation
threshold. Typically the percolation threshold ranges from 10-40% volume fraction of the
conductive filler (Jagur-Grodzinski, 2002). It has also been theoretically and experimentally
observed that for polymer blends including a dispersed metal(Bridge & Tee, 1990) or carbon
black(Carmona, 1989) phase, that the percolation threshold depends on the aspect ratio of the
filler and it decreases when the ratio of the length to the diameter increases(Munson-McGee,
1991). Gubbels et al. reported a percolation threshold as low as 3 weight percent carbon black
when it is localized in one phase of a binary blend consisting of 45% polyethylene and 55%
polystyrene(Gubbels, et al., 1995; Gubbels, et al., 1994). Then, they showed that localization of
carbon black at the interface of a co-continuous PE/PS blend exhibits a sharp reduction in the
percolation threshold value to 0.5 wt%. Blends consisting of conductive material prepared
slightly above the percolation threshold have low conductivity. The most difficult part is
preparing the conductive blend near the percolation threshold with reproducible conductive
results. Sumita et al.(Sumita, Sakata, Asai, Miyasaka, & Nakagawa, 1991) employed the
methodology of spreading coefficient to develop a criterion to determine the conditions of
localization of carbon black at the interface.
After the discovery of doped polyacetylene and its metal-like behavior, research shifted from
carbon black to intrinsically conductive polymer blends (ICP). By replacing the inorganic
semiconductors in conventional device structures with a semiconducting polymer, the basic
application of these materials has already been demonstrated(Burroughes, Jones, & Friend,
1988). They can store information and energy, and are capable of performing intelligent
functions. The most amenable conductive polymer to solution processing and meltprocessing(Paul,
Vijayanathan, & Pillai, 1999) is polyaniline(PANI). Investigations by
Macdiarmid et al.(Macdiarmid, et al., 1985), which resulted in the discovery of electrical
conductivity for the emeraldine salt of polyaniline, led to an explosion of interest in this
fascinating polymer. The emeraldine base of PANI is soluble in a few strong acids(Andreatta,
Cao, Chiang, Heeger, & Smith, 1988). Various dopants were utilized to induce meltprocessability
in PANI. The main disadvantage of all intrinsically conductive polymers including
PANI is its limited melt-processability. The addition of zinc compound to doped-PANI complex
99

allows for a method to tailor the processability of sulfonic acid-doped PANI(Hartikainen, et al.,
2001; Ruokolainen, et al., 2000). It has been observed that for polymer blends containing PANI,
a range of conductivity between 10 -10 S cm -1 to 10 -1 S cm -1 (melt processing) and 10 -10 S cm -1 to
10 S cm -1 (solution processing) can be achieved(Panipol, 2000).
The blending of PANI with a variety of classical polymers to improve the mechanical properties
of PANI and also to decrease the percolation threshold of the conductive polymers has been
extensively studied. Conductive binary polymer blends can demonstrate good mechanical
properties with conductivity varying from 10 −11 S cm -1 (almost insulating) to 300 S cm -1 . It
should be noted that different techniques including melt-blending, solution blending, in-situ
polymerization, and dispersion mixing have different influence on PANI morphology and
subsequently on properties. For instance, in the case of a PS/PANI blend, although melt-blending
results in a higher PANI percolation threshold than dispersion mixing, a significant improvement
in mechanical properties is achieved via the melt blending approach(Segal, et al., 2001). Most
studies of polymer blends containing PANI have been performed in solution and only a few
melt-blending studies have been reported(Ikkala et al., 1995; Narkis, Zilberman, & Siegmann,
1997; Shacklette, et al., 1993; Zilberman, et al., 1997). Polystyrene(PS)(Woo Jin, et al., 2003),
Cellulose acetate(Planes, et al., 1998), Polymethylmethacrylate(PMMA)(Morgan, et al., 2001),
Polyethylene(PE), Polypropylene(PP), Polyamide(PA)(Qinghua, et al., 2002),
Polyimide(PI)(Moon Gyu & Seung Soon, 2001), Polyvinyl chloride(PVC)(Chipara, et al., 1998),
Polyurethane(PU)(Rodrigues & Akcelrud, 2003), and a wide variety of thermoplastic
elastomers(Dong-Uk, et al., 2001; Ong, et al., 1997) have been blended with PANI using a
variety of techniques other than melt-blending(Ikkala, et al., 1995; Narkis, et al., 1997; Segal, et
al., 2001; Shacklette, et al., 1993; Zilberman, et al., 1997).
A conductivity value of 0.1 S cm -1 for 20 wt% of PANI is observed for PS/PANI blend prepared
by in-situ polymerization(Woo Jin, et al., 2003). Using a polymer solution blending approach,
Cao et al.(Cao, Smith, & Heeger, 1992) developed a processing method for a host polymer of
PMMA and polyaniline optimally doped with camphor sulfonic acid. Using this method the
value of electrical conductivity could be increased to 1 S cm -1 for loading levels of polyaniline
as low as 0.3% (vol/vol)(Yang, Y. Cao, P. Smith, & Heeger, 1993b). The solution blending of
PANI and its derivatives with poly(vinylidene fluoride)(PVDF) have attracted significant interest
due to some of the special properties of PVDF, namely, excellent mechanical properties,
100

chemical and weathering resistance, piezo-electric properties and good flexibility(Bliznyuk et al.,
2005; Privalko, et al., 2005). Fraysse et al(Fraysse, et al., 2000) were able to decrease the
percolation threshold of PANI sharply in a PANI/PMMA solution blend to 0.5 wt% of PANI.
Although the above methods using in-situ polymerization and polymer solution blending can
result in very low percolation thresholds of conductive polymer, they suffer from serious
disadvantageous such as poor mechanical properties. In addition, clearly the environmental and
health related issues associated with solvents and solvent removal are critical concerns. A melt
blending approach, from this point of view, is potentially a much more robust and
environmentally friendly approach.
In the case of melt-processing, Shacklette et al.(Shacklette, et al., 1993) reported percolation
thresholds in a range of 6-10% for PANI dispersed in polar polymers such as polycaprolactone
and poly(ethylene terphthalate glycol). Zilberman et al.(Zilberman, et al., 2000d) have reported
percolation thresholds of PS/PANI and linear low-density polyethylene(LLDPE)/PANI of about
30 wt% of PANI. They reported a lower percolation value of 20 wt% PANI for the binary blend
composed of CoPA/PANI.
Levon et al.(Levon, et al., 1993) were the first to suggest the concept of double percolation in a
ternary polymer blend consisting of a conductive polymer. This double percolated system was
comprised of a connected path within a connected path, the latter of which is conducting.
Zilberman et al.(Zilberman, et al., 2000d) observed a high-quality conductive PANI network in a
CoPA/LLDPE/PANI ternary blend and a poor-quality PANI network in a
(PS+DOP)/LLDPE/PANI blend. Narkis and coworkers(Haba, et al., 2000; Narkis, et al., 2000a;
Segal, et al., 2001)extended and developed these previous works by blending PANI with other
homopolymers via different methods mostly by using the methodology of a double percolated
morphology to decrease the percolation threshold and increase mechanical properties. They
showed that for several melt processed PANI/polymer blends, the conductivity results reveal a
percolation threshold of approximately 20%. That work also shows that, in ternary blends
consisting of PANI and two immiscible polymers, the PANI is preferentially located in one of
the constituting polymers. They found that PANI always prefers to locate in one phase rather
than situating at the interface(Zilberman, et al., 2000d). In all previous works performed on
PANI, the percolation threshold concentration of PANI was determined based on the onset of
increase in the conductivity as a function of composition. In most of those cases, the sample
101

conductivity at the percolation threshold point is between 10 -11 S cm -1 and 10 -9 S cm -1 . In order to
have higher conductivity values, higher concentrations of PANI are required.
This work reports on the development of onion morphology and multi-percolated morphologies
in HDPE/PS/PMMA/PVDF/PANI blends in which all phases are assembled hierarchically in
order. The overall objective is to lower the percolation threshold conductivity value of PANI to
the lowest possible value. The required thermodynamic conditions to obtain these structures will
be examined. The effect of the number of components, phase composition and the influence of
an interfacial modifier on the conductivity of the blends will be examined.
4.3 Experimental Methods
4.3.1 Materials
Commercial HDPE, PS, PMMA, processable PANI, and flouropolymer(PVDF) were examined
in this work. The main characteristics of the materials used in this study are represented in Table
4-1. Cyclohexane, acetic acid, dimethylformamide(DMF), and chromic acid were used to extract
or dissolve the selected phases.
PANI was purchased from Panipol Ltd, in the form of cylindrical compressed pellets with a
composition of about 25 weight % polyaniline salt and 75 weight % of a zinc compound to
modify its processability. The specific conductivity of this material is about 5 S cm -1 .
102

Table 4-1. Material Characteristics
Material Supplier Commercial
Code
Mw×10 -3
(g/mol)
η*× 10 -3 at
25 rad s -1 (Pa.s)
Density (g/cm 3 )
at 20 °C
103
Density (g/cm 3 )
at 200 °C
Poly(methyl methacrylate) Aldrich - 12 0.04 a 1.19 1
Polystyrene Dow 615APR 290 1.5 1.04 0.97
High density Polyethylene Dow - 79 0.72 0.98 0.85
Polyaniline Panipol CX - 0.11 b - 1.07
Polyvinylidene fluoride Arkema Kynar Flex - 1.4 c - 1.6
PS-co-PMMA (40% Styrene) - - 100-150 34 d - 0.98
a reported by Reignier et al.(Reignier & Favis, 2000)
b viscosity measured at 180°C
c zero shear viscosity of PVDF is 902000 Pa.s at 190°C calculated by Carreau-Yasuda model
d zero shear viscosity (Pa.s) at 190°C
4.3.2 Rheological Analysis
The rheological characterization of the neat polymers was measured using a Bohlin constant
stress rheometer (CSM) in the dynamic mode. PS, PMMA, HDPE, and PVDF are known to
follow the Cox-Mertz equation(Cox & Merz, 1958), consequently, the applied frequency can be
considered as the shear rate and the complex viscosity as the steady shear viscosity.
Measurements were carried out at 190 ºC under a nitrogen atmosphere. The complex viscosity of
the pure homopolymers are plotted as a function of frequency in Figure 4-2. The viscosity of
homopolymers at 25 s -1 are reported in Table 4-1. All polymers show shear-thinning behavior
and the Carreau-Yasuda model(Yasuda, Armstrong, & Cohen, 1981) was used to determine the
zero-shear viscosity of PVDF.

Complex voscosity (Pa.s)
1,0E+06
1,0E+05
1,0E+04
1,0E+03
1,0E+02
1,0E+01
1,0E+00
Figure 4-2. Complex viscosity as a function of frequency at 190ºC for the homopolymers. The
complex viscosity of PANI is at 180ºC.
4.3.3 Sample Preparation
0,01 0,1 1 10 100 1000
Frequency (rad s -1 )
Various blends were prepared via melt-blending under a flow of dry nitrogen in a 30mL
Brabender operating at a set temperature of 200 °C and 50 rpm. The maximum shear rate at this
speed is close to 25 s -1 .(Philippe Lavalle, et al., 2004) The real temperature achieved by the end
of mixing experiment was approximately 185°C. A value of 0.7 fill factor was selected and thus
after converting the volume to the mass and weighing the material, the mixing chamber was
filled to 70% of its total volume. A concentration of 0.2 weight percent of Irganox antioxidant
supplied by CIBA was added to the mixture to reduce the thermal oxidation of the components.
All the blends were mixed for 8 min. After average 2 min mixing time the torque achieves a
constant plateau value. After mixing, the samples were immediately cut from the mass and
quenched in a cold water bath to freeze-in the morphology.
HDPE
In order to produce disk-shaped samples of 2.5cm diameter and 1.2mm thickness for
conductivity testing, a hot press was employed. The press is heated to 200°C and the mould is
PS
PVDF
PANI
104

filled with blend pellets. To facilitate the de-molding of the material, two sheets of Teflon are
inserted between the plaques. The total molding cycle takes 8min under a nitrogen purge gas. A
large cold press was employed after hot press to decrease the temperature sharply to avoid
annealing of the sample.
4.3.4 Solvent Extraction
A solvent extraction/gravimetric method was used to obtain quantitative data on the extent of
continuity of the phases. Selective solvents were used to extract the specimens of various masses
at room temperature. The volume of the components before and after extraction is measured by
weighing the sample and converting the weight to volume(Equation 4-3). As a primary
advantage, solvent extraction is an absolute measurement and is also a straightforward technique
to detect the existence of co-continuous microstructures when the components are soluble in
specific solvents.
Equation 4-3.
% Cocontinuity
=
∑Vinitial − ∑
n n
Where Vinitial and Vfinal are the volume of one component present in the sample before and after
extraction calculated by weighing the sample and converting it to the volume. The degree of
continuity represents the fraction of a phase that is continuous. Samples in which each phase has
a degree of continuity of 1.0 are completely continuous. The reported value is the average of
several samples and the average error for high continuity levels is ± 5% and for low continuity
levels is ± 3%. Cyclohexane, acetic acid, and chromic acid were utilized as selective solvents for
polystyrene, polymethyl methacrylate, and polyaniline, respectively and DMF extraction was
performed to dissolve PS, PMMA, and PVDF altogether.
∑
n
V
initial
V
final
105

4.3.5 Characterization of Phase Morphology
4.3.5.1 Microtoming and Scanning Electron Microscopy (SEM)
The specimens were cut and microtomed to a plane face under liquid nitrogen using a microtome
(Leica-Jung RM 2065) equipped with a glass knife and a cryochamber type (LN 21). After the
appropriate chemical treatment at room temperature with selective solvents (acetic acid for
PMMA, cyclohexane for PS, chromic acid for PANI, and DMF for all three of PVDF, PMMA,
and PS) to remove one or several of the components, the sample surface was coated with a
gold/palladium by plasma deposition. A JEOL JSM 840 scanning electron microscope, operated
at a voltage of 10 keV to 12 keV, was used to obtain photomicrographs of the sample surface.
4.3.5.2 Focused Ion Beam (FIB), Atomic Force Microscopy (AFM)
A FIB/AFM technique was used to examine the PE/PS/PMMA blend. After cryomicrotoming of
the specimens using a glass knife to obtain perfectly flat surface, samples were coated with a
gold-palladium alloy. The surface of the samples was then treated and etched using a Hitachi
focused ion beam FIB-2000A operated at 30 keV gallium beam. FIB uses a focused beam of
gallium ions which are accelerated to energy of 5-50 kiloelectronvolts. Using an electrostatic
lens, ion beam can be focused on a very small spot, resulting in a high resolution because of the
small emitting area. In this work, a 0.8 nA beam current and a dwell time at 3 μs was applied in
order to remove approximately 3-4 μm of the surface. Milling was carried out parallel to the
observed surface. Since each polymer component has a different interaction with the gallium
beam, this approach induces topological differences and hence increases the contrast between
components. The milled surface of the specimen was then examined by a scanning probe
microscope dimension (AFM) with a Nanoscope IIIa controlled in topological mode. The atomic
force microscope measures topography with a force probe. Silicon tips with spring constant of 40
N/m and a resonant frequency of around 300 kHz was employed.
106

4.3.6 Conductivity Measurements
DC electrical measurements were performed through the vertical thickness of the substrate at
ambient temperatures using a Keithley electrometer model 6517. The sample is placed between
two sample holders and a DC voltage is applied using two point probe equipment. The setup
works at a voltage range of 0-1000V and an ammeter range of 0-20mA. To standardize equal
pressure for all samples, a 5lb load was applied on the sample holders. Cylindrical specimens
with a thickness of 1.2mm and a radius of 1.25cm were tested in dry air. Although the top and
bottom surfaces of the samples are flat, two graphite sheets covered the surface of the sample in
order to ensure good contact of the sample surface and electrodes. The electrical conductivity of
the samples varied over a wide range from 10 -12 S cm -1 to 10 -1 S cm -1 . The average % error in the
conductivity measurements is ± 0.2 orders of magnitude.
4.4 Results and Discussions
4.4.1 Reducing the Percolation Threshold of PANI in a Multi-component
Polymer Blend
The principal objective of this work is to control the morphology of a multi-component polymer
blend in order to significantly reduce the percolation threshold of the conductive PANI
component. The first question to be addressed is where does the PANI need to be located in the
blend system?
In recent work, Zhang et al.(J. Zhang, et al., 2007) demonstrated that the percolation threshold of
one component could be reduced to values as low as three percent by preparing a ternary blend
and situating that phase at the interface of two other co-continuous phases. In Figure 4-3a, it is
shown that polystyrene situates at the interface of PE and PMMA and the ordering of the phases
can be directly related to Harkins spreading theory. In Figure 4-3a, the identification of the
phases can be clearly seen by the topographical heights induced by FIB etching and subsequently
quantified by AFM analysis in the topographical mode. Previous work showed that PMMA is
107

less etched than HDPE which is less etched than PS(Virgilio, Favis, Pepin, Desjardins, &
L'Esperance, 2005a). The positive spreading coefficient of PS over PMMA (λPS/PMMA) with a
value of 2.6 mN/m predicts the development of a thermodynamically stable PS layer between the
PE and PMMA phases and is a direct result of the very high interfacial tension of PMMA and
HDPE(γHDPE/PMMA = 8.4 mN/m).
a)
PS
HDPE
PE
PS
PMMA
PMMA
PMMA
Distance (µm)
PS
PE
b)
Figure 4-3. a) FIB-AFM image of the composite-droplet morphology of 30/10/60
HDPE/PS/PMMA showing PS as a layer separating HDPE and PMMA. Scan size is 5μm×5μm.
The bar to the right of the FIB/AFM micrograph indicates the colours associated with the
topographical height in nm. The white line in the image indicates the section analyzed below,
and b) SEM micrograph of a 45/10/45 HDPE/PVDF/PANI blend after extraction of the PVDF
phase by DMF. The voids show that PVDF was present as a layer separating HDPE and PANI.
Table 4-2 represents the surface tension and polarity of the various polymers being used in the
harmonic mean equation(Wu, 1982) for interfacial tension. Table 4-3 shows a range of
interfacial tension values of polymer pairs, including PANI with other polymers, estimated using
the harmonic mean equation. A number of these interfacial tensions were also measured
108

experimentally in this research group and they show a good correlation with the harmonic mean
data(Table 4-3).
Table 4-2. Surface Tension, Polar Contribution, Dispersive Contribution, Polarity, and Rate of
Surface Tension per Temperature for Homopolymers
Polymers γ(mN/m)
at 20°C
γ(mN/m)
at 200°C
γ p /(mN/m)
at 200°C
γ d (mN/m)
at 200°C
Polarity,
X p
(-dγ/dT)
(mN/m-deg)
HDPE a 35.7 25.4 0 25.4 0 0.057
PS a 40.7 27.7 4.6 23.1 0.167 0.072
PMMA a 41.1 27.4 7.7 19.7 0.28 0.076
PVDF b 33.2 22.9 8.7 14.2 0.376 0.057
PANI c 55.6 40.3 24.6 15.7 0.612 0.085
a From ref. (Wu, 1982), b From ref. (Baradie & Shoichet, 2003), c Measured by contact angle method
Table 4-3. Theoretical and Experimental Interfacial Tension for Polymer Pairs
Theoretical data(Harmonic mean equation) γ (mN/m) Experimental dataγ (mN/m)
PS/HDPE 4.7 PMMA/HDPE 8.4 PS/HDPE a 5.1
PMMA/PS 1. 1 PANI/HDPE 26.9 PMMA/PS a 2.4
PVDF/HDPE 11.9 PVDF/PMMA 1 PMMA/HDPE b 8.5
PVDF/PS 3.4 PVDF/PANI 7.7
PANI/PMMA 9.3 PANI/PS 15.1
a From ref. (Reignier & Favis, 2000), b Measured by breaking thread method
109

The harmonic mean equation approach is a classic approach for estimating the interfacial tension
of polymer pairs and has been used widely in a number of studies(Wu, 1982). It can be seen
that, compared to other polymer pairs, mixtures with PANI result in exceptionally high
interfacial tensions. It thus follows from spreading theory that PANI will not be driven to
encapsulate other polymers in a multi-component blend, rather it will tend to be encapsulated by
virtually every other polymer species. For example, in a ternary HDPE/PVDF/PANI blend, a
positive spreading coefficient value of 7.3 mN/m for λPVDF/PANI predicts encapsulation of PANI
by PVDF as confirmed in Figure 4-3b. In that figure, the voids left by the extracted PVDF can be
clearly seen to show that PVDF formed an interlayer between HDPE and PANI.
By increasing the number of components in a multi-percolated structure, the percolation
threshold of all phases sharply decreases. Thus, a unique approach to achieve a low percolation
threshold conductive PANI device would be to prepare a multiple percolated blend system with
PANI as the innermost phase. In this work we will examine the progressive morphological
development of highly continuous multi-component structures by first examining the
development of multi-encapsulated ternary to quinary droplets which are referred to here as
onion morphologies. Then, the required conditions for transformation of the onion morphology
to continuous multi-percolated morphologies and the conductivity of those systems are
examined.
4.4.2 Self-Assembled Onion Morphology and the Hierarchical Ordering of
Phases
Typically, in the literature, a core/shell morphology is used to describe a droplet comprised of
one phase encapsulated by another single phase. Here, we use the term onion morphology,
schematically shown in Figure 4-4, to describe a multi-phase, multi-encapsulated dispersed
droplet structure. In other words, a polymeric droplet phase comprised of layers of different
polymers assembled concentrically. Figure 4-4 shows schematically the progression from a
matrix-droplet(Figure 4-4a) to a core/shell(Figure 4-4b) and finally two cases of onion
morphology(Figures 4-4c and 4-4d).
110

a) b) c) d)
Figure 4-4. Schematic illustration of various encapsulated structures, a) dispersed phase B in
matrix A, b) core-shell morphology B/C in matrix A, c) quaternary onion morphology B/C/D in
matrix A, and d) quinary onion morphology B/C/D/E in matrix A.
The thermodynamic conditions to obtain such onion structures for a given n-component blend is
that the spreading coefficients of each of three selected components out of n components should
satisfy the complete wetting case in Harkins equation. At the same time, the order of phases in
each selected ternary blend should be consistent with the order of the phases situated in the
multi-blend. This implies that for a core/shell structure, as there are only three phases in the
blend system, only one set of Harkins equation needs to be examined. In the case of a quaternary
blend system with an onion structure, a combination of three out of four which is equal to four
systems of Harkins equation should be evaluated. Finally in this study, ten systems of Harkins
equation (equivalent to three out of five components) should be evaluated for a quinary blend
system, which in the present work is comprised of HDPE, PS, PMMA, PVDF, and PANI. Both
the interfacial tensions and the spreading coefficients for these ten binary pairs are shown in
Tables 4-3 and Table 4-4, respectively.
The first four ternary blends in Table 4-4 indicate that in a quaternary blend comprised of an
HDPE matrix with PS, PMMA, and PANI, the order of components from outer to inner should
be: HDPE|PS|PMMA|PVDF. Using the data from Tables 4-3 and 4-4 it is also predicted that
various other quaternary mixtures should assemble hierarchically as follows:
HDPE|PS|PMMA|PANI (see spreading coefficients 1,4,5 and 8 from Table 4-4);
HDPE|PMMA|PVDF|PANI (spreading coefficients 3,5,7 and 10); HDPE|PS|PVDF|PANI
111

(spreading coefficients 2, 6, 7, and 9); and PS|PMMA|PVDF|PANI (spreading coefficients 4, 8,9,
and 10. Finally, from the above, the quinary blend should have the hierarchical ordering of
HDPE|PS|PMMA|PVDF|PANI.
Table 4-4. Spreading Coefficients in Different Ternary Blends and Predicted Order of Phases in
Blends
Spreading coefficients
λ (mN/m)
Ternary blend and order
of phases in the blend
Spreading coefficients
λ (mN/m)
112
Ternary blend and order
of phases in the blend
1 λPS/PMMA = 2.6 mN/m HDPE|PS|PMMA 6 λPS/PANI = 7.1mN/m HDPE|PS|PANI
2 λPS/PVDF = 3.8 mN/m HDPE|PS|PVDF 7 λPVDF/PANI = 7.3 mN/m HDPE|PVDF|PANI
3 λPMMA/PVDF = 2.5 mN/m HDPE|PMMA|PVDF 8 λPMMA/PANI = 4.7 mN/m PS|PMMA|PANI
4 λPMMA/PVDF = 1.3 mN/m PS|PMMA|PVDF 9 λPVDF/PANI = 4 mN/m PS|PVDF|PANI
5 λPMMA/PANI = 9.2mN/m HDPE|PMMA|PANI 10 λPVDF/PANI = 0.6 mN/m PMMA|PVDF|PANI
After a theoretical determination of the order of all phases, morphology experiments for the
binary, ternary, and quaternary blends comprised of the previously mentioned components
confirm the theoretical ordering. Some examples are shown in Figure 4-5. The binary blend of
80/20 PMMA/PS with matrix-droplet morphology is shown in Figure 4-5a. A ternary blend of
30/10/60 HDPE/PS/PMMA blends has already been shown in Figure 4-3a and confirms the
situation of PS at the interface of PE and PMMA which corresponds to the positive value of
λPS/PMMA(2.6 mN/m) (Case 1 from Table 4-4) as already discussed. Another ternary blend of
60/20/20 HDPE/PMMA/PVDF is shown in Figure 4-5b. In that case the PMMA has been
selectively extracted showing clearly a core/shell structure where PMMA is the shell and PVDF
is the core. This HDPE/PMMA/PVDF ternary blend corresponds to λPMMA/PVDF(2.5 mN/m) (Case
3 from Table 4-4) which predicts an encapsulation of PMMA about PVDF. The morphology
observed for a quaternary blend of 50/20/10/20 HDPE/PS/PMMA/PVDF, shown in Figure 4-5c,
is also consistent with the theoretical ordering discussed above. Figure 4-5c shows the onion-like

microstructure of the quaternary blend after selective dissolution of PMMA. The selective
dissolution of PMMA allows for the clear identification of the other phases. There is a core of
PVDF encapsulated in a first shell of PMMA, encapsulated in second shell of PS, all of which
are encapsulated by the HDPE matrix.
a)
c)
HDPE
Figure 4-5. SEM micrograph of a) matrix-droplet morphology of 80/20 PMMA/PS, b) PVDF
core-PMMA shell morphology in an HDPE matrix for 60/20/20 HDPE/PMMA/PVDF (PMMA
extracted), and c) onion morphology in an HDPE matrix for 60/13/13/13
HDPE/PS/PMMA/PVDF (PMMA extracted)
Theoretically, the addition of PANI as a fifth component to the quaternary blend above, creates a
five-component onion droplet structure with PANI as the inner phase and HDPE as the matrix.
b)
PMMA
PMMA
PS
PVDF
PVDF
HDPE
113

Since it is difficult to isolate the position of PANI in a 5 component blend system, a ternary
blend of HDPE/PVDF/PANI, already shown in Figure 4-3b was prepared. The morphology of
this blend, after selective extraction of the PVDF, clearly shows the localization of PVDF at the
interface of HDPE and PANI which is a result of the very high spreading coefficient of PVDF
over PANI, λPVDF/PANI (7.3 mN/m). These data on the PE/PVDF/PANI blends as well as the
discussion related to the other ternary and quaternary blend systems above would tend to confirm
the ordering of phases in the quinary system as HDPE|PS|PMMA|PVDF|PANI as was predicted
from Harkins spreading theory.
Although detecting the position of each phase in a five-component onion structure by
microscopy techniques is very difficult, the morphology of 50/10/20/10/10
HDPE/PMMA/PS/PVDF/PANI blend after selective extraction of one or several phases is
represented in Figure 4-6. Figure 4-6a depicts the continuous structure of HDPE after extraction
of all other phases. Empty spherical areas observed in the matrix of HDPE represent onion
domains containing various phases. The selective extraction of PS (Figure 4-6b) and PMMA
(Figure 4-6c) phases from the blend verifies them as two layers of an onion. Figure 4-6d shows
the same blend where phases PS and PMMA have been extracted together demonstrating a wider
empty area, indicating the successive situation of PS and PMMA phases. The extracted area in
Figure 4-6d separates two major phases which are the matrix of HDPE and droplet-in-droplet of
PANI-in-PVDF. All of the above confirms the hierarchical ordering of
HDPE|PS|PMMA|PVDF|PANI in the quinary blend.
114

a) b)
c) d)
Figure 4-6. SEM micrographs of 50/10/20/10/10 HDPE/PMMA/PS/PVDF/PANI after extraction
of (a) all phases except HDPE, (b) PS by cyclohexane (c) PMMA by acetic acid, and (d) PS and
PMMA by combining cyclohexane and acetic acid
4.4.3 Generating Highly Continuous (Multiple-Percolated) Quaternary
Polymer Blends
In this part of the work the composition range required to achieve multiple percolation for
quaternary HDPE/PS/PMMA/PVDF blends, where all phases are continuous and percolated
through the system, will be determined. In such a case, not only does the typical percolation
threshold of the multiphase blend decrease due to the increased number of components, but also
115

the specific percolation threshold of each of the individual phases (middle, inner, and outer)
significantly decreases.
The most critical phases in HDPE/PS/PMMA/PVDF are the first(HDPE) and last(PVDF) phases.
Changing the ratio of these two phases, while maintaining the concentration of the middle phases
controls the evolution of the morphology from an onion structure to a multi-percolated one. It
has been shown in previous work from this laboratory(Reignier, et al., 2003) that composite
droplets with a core/shell morphology can experience coalescence and still maintain the
hierarchical order after coalescence. Figure 4-7a-e, which shows quaternary
HDPE/PS/PMMA/PVDF blends of concentrations of 60/13/13/13(onion morphology),
50/17/17/17, and 30/15/15/40(multi-percolated morphology), illustrates that the hierarchical
order is maintained as the morphology transits from an onion structure to more elongated
structures through to a multi-percolated one. Figure 4-7e shows unambiguously, after the
selective extraction of the PMMA phase, that highly ordered multipercolated morphology is
obtained for 30HDPE/15PS/15PMMA/40PVDF with a hierarchical ordering of
HDPE/PS/PMMA/PVDF.
116

a) b)
PVDF
PS
PMMA
c) d)
e)
HDPE
PS
PVDF
HDPE
Figure 4-7. SEM micrographs and schematics of HDPE/PS/PMMA/PVDF after extraction of
PMMA by acetic acid for a) and b) onion morphology in an HDPE matrix for 60/13/13/13, c)
and d) more elongated morphology of 50/17/17/17, and e) triple-percolated morphology of
30/15/15/40 (microtomed surface). The voids show that PMMA is present as a layer separating
PS and PVDF.
117

a) b)
c) d)
e)
Figure 4-8. SEM micrographs of various quaternary HDPE/PS/PMMA/PVDF blends after
microtoming and extraction of the PMMA phase; (a) 10/15/15/60, (b) 20/15/15/50, (c)
30/15/15/40, (d) 40/15/15/30, (e) 50/15/15/20, (f) 60/15/15/10.
f)
118

Uniform layers of PS and PMMA situated in the middle of a quaternary blend of
HDPE/PS/PMMA/PVDF are found at concentrations levels of 15% for each respectively. In
order to determine the range of multi-percolated regions in the quaternary
HDPE/PS/PMMA/PVDF blend, the composition ratio of HDPE and PVDF is varied from
10/15/15/60 HDPE/PS/PMMA/PVDF to 60/15/15/10 HDPE/PS/PMMA/PVDF while the
composition of PS and PMMA is maintained at 15% (Figure 4-8).
Figure 4-8 illustrates the morphologies obtained and in all cases PMMA has been extracted by
acetic acid to distinguish it from HDPE, PS, and PVDF in the blend. Hence, PMMA can be
recognized as void areas separating the continuous parts of the PS layer attached to HDPE and
continuous PVDF. When the composition of HDPE is as low as 10% and 20%, droplets of
HDPE encapsulated by the PS phase are detected, and both are encapsulated by PMMA in a
matrix of PVDF(Figure 4-8a and Figure 4-8b). The opposite behavior is encountered when the
PVDF phase has concentrations as low as 10% and 20%, demonstrating droplets of PVDF
encapsulated by PMMA phase where both are encapsulated by PS within a HDPE matrix (Figure
4-8e and Figure 4-8f). A comparison of 10/15/15/60 HDPE/PS/PMMA/PVDF (Figure 4-8a) and
60/15/15/10 HDPE/PS/PMMA/PVDF(Figure 4-8f) reveals that, although the order of phases
does not change, the onion morphology is reversed depending on whether the inner phase or
outer phase is the matrix. As the concentration of HDPE or PVDF further increases and reaches
30%, droplets of those phases coalesce and continuous structures of the respective phase forms.
Blends of 30/15/15/40 HDPE/PS/PMMA/PVDF and 40/15/15/30 HDPE/PS/PMMA/PVDF show
two continuous phases of HDPE and PVDF while continuous PS and PMMA phases are situated
at the interface (Figure 4-8c and Figure 4-8d).
In order to find the precise region of multi-percolation quantitatively, solvent extraction followed
by a gravimetric measurement was used to obtain the percentage continuity of each phase as a
function of volume fraction of HDPE in the quaternary blend of HDPE/PS/PMMA/PVDF
(Figure 4-9). Note that in Figure 4-9 the concentrations of PS and PMMA are both held at 15%,
while the combined concentration of HDPE plus PVDF is always held at 70%. In Figure 4-9, the
x axis refers to the % concentration of HDPE in HDPE plus PVDF. Thus 100% HDPE refers to
the 70HDPE/15PS/15PMMA blend while 50% HDPE refers to the
35HDPE/15PS/15PMMA/35PVDF blend.
119

Phase continuity (%)
100
90
80
70
60
50
40
30
20
10
0
Continuity of PMMA Continuity of PMMA+PS Continuity of PS
(a) (b) (c) (d) (e) (f)
0 20 40 60 80 100
Volume fraction of HDPE(HDPE+PVDF = 70% of total)
Figure 4-9. Continuity level of PS, PMMA, and PS+PMMA obtained by solvent
extraction/gravimetry for 6 blends comprised of HDPE/PS/PMMA/PVDF; (a) 10/15/15/60, (b)
20/15/15/50, (c) 30/15/15/40, (d) 40/15/15/30, (e) 50/15/15/20, (f) 60/15/15/10.
Selective extraction associated with gravimetric measurements was carried out to obtain the
continuity levels of PS, PMMA, and PS+PMMA in order to determine the concentrations where
the phases are continuous. The continuity of HDPE and PVDF were estimated based on SEM
photomicrographs at various compositions. Gravimetric results exhibit a low level of continuity
(ranging from five to fifteen percent) for all phases in a 10/15/15/60 HDPE/PS/PMMA/PVDF
onion morphology. Increasing the concentration of HDPE to 20 parts in the blend with PVDF at
50 (28.6% HDPE on the x-axis) yields a hierarchical-layered morphology with an HDPE core
where the continuity of PS and PMMA reaches levels of 30% and 43%, respectively. Further
increasing the HDPE concentration results in a maximum continuity of all phases for blends of
30/15/15/40 HDPE/PS/PMMA/PVDF and 40/15/15/30 HDPE/PS/PMMA/PVDF (Figure 4-9). If
120

the concentration of HDPE is increased further, an inverse behavior is observed where PVDF
becomes the core and HDPE becomes the matrix. In this case, for the blend containing
50HDPE/15PS/15PMMA/20PVDF, the continuity of middle phases including PS and PMMA
becomes 15% and 30%, respectively. Figure 4-9 shows that when the concentration of HDPE is
increased up to 60 parts in the blend, the continuity of PS and PMMA becomes less than five
percent.
4.4.4 Generating Highly Continuous Quinary Polymer Blends
In this part of the work, the addition of a fifth component to the highly continuous quaternary
blends ranging from 30/15/15/30 to 40/15/15/30 HDPE/PS/PMMA/PVDF, selected from the
previous section, is examined.
Ten percent PANI is added to a quaternary blend of HDPE/PS/PMMA/PVDF within the region
of multi-percolation resulting in the formation of a five-component blend of 30/15/15/30/10
HDPE/PS/PMMA/PVDF/PANI(Figure 4-10). Figure 4-10a shows the continuous HDPE phase
after the removal of all other phases. The selective extraction of PMMA and PS reveals a
perfectly homogeneous continuous network for those phases in the blend (Figure 4-10b and
Figure 4-10c). The extraction of both PS and PMMA phases together separates the multi-blend
into two inter-diffused continuous structures where one of them is a continuous HDPE phase and
the other is a continuous PANI and PVDF phase (Figure 4-10d). It also illustrates that continuous
networks of PS and PMMA are adjacent to each other. In Figure 4-10e, the network of PANI
which has been extracted by chromic acid is clearly detected.
In order to unambiguously illustrate that PANI forms a network morphology, blends of
25/25/25/25 PS/PMMA/PVDF/PANI and 15/20/15/25/25 PS/PS-co-
PMMA/PMMA/PVDF/PANI are prepared. The influence of the addition of a PS-co-PMMA
copolymer on the segregation of phases is discussed in a following section in detail. After the
extraction of all phases except PANI and followed by a freeze-drying process, a network of
PANI with very thin walls remains, as represented in Figure 4-11. Freeze-drying is carried out to
prevent of shrinkage of the PANI network.
121

a) b)
c)
d) e)
Figure 4-10. SEM micrographs of the multi-percolated 30/15/15/30/10 blend comprised of
HDPE/PS/PMMA/PVDF/PANI after extraction of (a) all phases except HDPE, (b) PMMA by
acetic acid (c) PS by cyclohexane, (d) PS and PMMA by combining cyclohexane and acetic acid,
and (e) PANI by chromic acid
122

a) b)
c)
e)
Figure 4-11. SEM micrographs of a), b) continuous PANI network of quaternary 25/25/25/25
PS/PMMA/PVDF/PANI blend after extraction of all phases by DMF followed by a freeze drying
process, and c, d, e) continuous PANI network derived from a 15/20/15/25/25 PS/PS-co-
PMMA/PMMA/PVDF/PANI blend after extraction of all phases by DMF followed by freeze
drying process. The thickness of the PANI network walls are shown in Figure 4-11e.
d)
123

a)
c)
Figure 4-12. a) Schematic representation of quinary multi-percolated morphology, the yellow
phase is PANI, the blue phase is HDPE, and all other phases are located between them, b)
schematic of a PANI network obtained after removal of all phases except PANI, and c) a 3D
image constructed from 2D SEM images of the interconnected PANI network in the quinary
15/20/15/25/25 PS/PS-co-PMMA/PMMA/PVDF/PANI blend after extraction of all other phases.
This figure clearly confirms that in a triple-percolated structure with PANI as the inner phase,
PANI will spread. As shown in Figure 4-11e, the PANI network in a multi-percolated structure
has branches with very thin walls. The range of thicknesses of the wall in this structure range
from a few hundred nanometers to 700 nanometers. Another important point that should be taken
into account is the actual content of pure PANI in the blends. The PANI used in this work is
b)
124

melt-processable and contains 25 wt% pure PANI and 75 wt% zinc compound. This indicates
that the real volume of pure PANI present in the blends is less than reported volume fraction in
these blends.
A schematic representation for such a 5-component multi-percolated structure is shown in Figure
4-12a. The yellow phase represents PANI which, after removal of all other phases by selective
solvents, demonstrates an interconnected network as schematically shown in Figure 4-12b.
Figure 4-12c illustrates the 15/20/15/25/25 PS/PS-co-PMMA/PMMA/PVDF/PANI blend after
extraction of all phases except PANI, clearly indicating that a complete PANI network is
attained.
4.4.5 The Effect of Diblock Copolymer on the Morphology and Conductivity
of Multiple-Percolated Blend Systems
The addition of a PS-co-PMMA copolymer to a PS/PMMA/PVDF/PANI multi-percolated
structure results in a decrease of the size of the PS and PMMA phases. Such a decrease in the
size of two adjacent phases results in corresponding smaller phase sizes for the other phases
involved in the multi-encapsulated structure. This can be used as an additional parameter to
control morphology and conductivity.
Figure 4-13 depicts two multi-percolated quaternary blends, one without modifier (30/30/30/10
PS/PMMA/PVDF/PANI) and the other with modifier (20/20/20/30/10 PS/PS-co-
PMMA/PMMA/PVDF/PANI). In both cases, the composition of all the phases is identical.
Figure 4-13a and Figure 4-13b compare samples with and without a copolymer, respectively,
after extraction of both PS and PMMA phases using cyclohexane and acetic acid, respectively.
The addition of PS-co-PMMA to the multi-percolated PS/PMMA/PVDF/PANI blend clearly
reduces the sizes of all the phases.
Figures 4-13c and 4-13d examine the effect of added copolymer on the phase size of PANI alone
where PANI is removed using chromic acid. These micrographs show that the PANI phase size
is significantly smaller after addition of PS-co-PMMA. Thus, the addition of a copolymer
125

specific to two adjacent phases in a multi-encapsulated system results in a reduction of all phase
sizes in the encapsulated system.
Figure 4-11 shows the effect of copolymer more clearly on the PANI network structure after
removal of all other components followed by a freeze drying step. Two blends are considered,
one without copolymer, 25/25/25/25 PS/PMMA/PVDF/PANI (Figure 4-11a and b), and one
containing copolymer, 15/20/15/25/25 PS/PS-co-PMMA/PMMA/PVDF/PANI (Figure 4-11c, d,
and e). The micrographs demonstrate that, in the presence of copolymer, the PANI network is
finer and more uniform with homogeneous branches distributed throughout the sample.
a) b)
c) d)
Figure 4-13. SEM micrograph of a) 30/30/30/10 PMMA/PS/PVDF/PANI blend after extraction
of PS and PMMA, b) 20/20/20/30/10 PS/PS-co-PMMA/PMMA/PVDF/PANI after extraction of
PS and PMMA, c) 30/30/30/10 PMMA/PS/PVDF/PANI blend after extraction of PANI by
chromic acid, and d) 20/20/20/30/10 PS/PS-co-PMMA/PMMA/PVDF/PANI after extraction of
PANI by chromic acid
126

One question that should be addressed is whether the addition of a PS-co-PMMA interfacial
modifier would be expected to theoretically affect the hierarchical order in these blend systems.
Table 4-3 shows a very low value of interfacial tension of γPS/PMMA=1.1 mN/m. If after addition
of copolymer, γPS/PMMA decreases to the lowest possible value close to zero, all ternary blends
comprised of both PS and PMMA in the blend still show positive spreading coefficient values.
These results imply that the positive values of λPS/PMMA in HDPE|PS|PMMA, λPMMA/PVDF in
PS|PMMA|PVDF, and λPMMA/PANI in PS|PMMA|PANI result in three complete wetting cases
with the same order of phases as before the addition of copolymer.
4.4.6 The Effect of the Number of Components of Multi-Percolated
Structures on Conductivity
In this section the effect of the number of the components in multi-percolated structures on the
conductivity, where PANI phase situated as the innermost phase, is investigated as a function of
volume fraction of PANI (Figure 4-14).
The percolation threshold for a conductive material is the concentration at which the first
connected pathway of conductive polymer forms in the blend. Conductivity data thus provides a
tool as powerful as solvent gravimetry to determine percolation thresholds for a conductive
substance. In a blend containing conductive material can be used for both conductivity and
continuity. It implies that the concentration at which the first connected pathway of conductive
polymer in the blend forms, both conductivity and continuity values sharply increase resulting in
percolation threshold. Hence, conductivity percolation threshold corresponds to continuity
percolation threshold of sample. In an onion morphology with the PANI phase located at the core
of a dispersed phase with low connectivity, the conductivity would be expected to be low. Figure
4-14 shows samples with multi-percolated structures constituting various numbers of
components. The concentration of each sample is represented in Table 4-5.
127

Table 4-5. Number of Components and Compositions of Samples Shown in Figure 4-14
Sample Number of components PANI PS PVDF PMMA HDPE PS-co-PMMA
A 3 5 50 ─ 45 ─ ─
B 4 5 15 45 ─ 35 ─
C 5 5 20 35 20 20 ─
D 6 5 16 20 15 35 9
E 3 10 45 ─ 45 ─ ─
F 4 10 30 30 30 ─ ─
G 5 10 20 20 20 30 ─
H 5 10 20 30 20 ─ 20
I 3 20 40 ─ 40 ─ ─
J 5 20 20 20 20 20 ─
K 6 20 15 20 12 20 13
L 4 25 25 25 25 ─ ─
M 5 25 15 25 15 ─ 20
N 3 33.3 ─ 33.3 ─ 33.3 ─
O 2 50 50 ─ ─ ─ ─
128

Conductivity(S/cm)
Ternary Blend
1,0E-01
Quaternary blend Quinary blend (Quaternary or Quinary blend)/Copolymer
1,0E-02
1,0E-03
H
G
K
J
M
L
N
O
1,0E-04
1,0E-05
1,0E-06
1,0E-07
1,0E-08
1,0E-09
1,0E-10
1,0E-11
1,0E-12
D
C
B
A
F
E
I
129
0 5 10 15 20 25 30 35 40 45 50 55
Volume Fraction of PANI(%)
Figure 4-14. Conductivity of multi-percolated blends of ternary, quaternary, quinary, as well as
interfacially modified quaternary and quinary systems as a function of the volume fraction of
PANI. Note that point O is a co-continuous binary blend of PS and PANI.
In Figure 4-14, samples A, B, C, and D represent ternary, quaternary, five-component, and sixcomponent
polymer blends all containing five percent PANI. In all samples, the composition of
all phases, other than PANI, have been selected to be in the multi-percolated region(Table 4-5).
Sample (A) is a ternary blend of 5/45/50 PANI/PMMA/PS in which PMMA is at the interface.
Due to the low amount of PANI (5%) in this sample, droplets of PANI are encapsulated by the
PMMA phase. The very low conductivity value of almost 10 -11 S cm -1 indicates that no
connected conductive pathway forms in this sample. In order to spread the PANI at least four
components in the multi-percolated structure are required. In sample (B), five percent of PANI is
present in a quaternary blend with a composition of 35/15/45/5 HDPE/PS/PVDF/PANI. In this
blend, due to the geometrical restrictions imposed on PANI, a continuous PANI phase forms a

network and the percolation threshold of PANI is achieved as seen by the conductivity value of
3.5×10 -7 S cm -1 . Although the percolation threshold is defined as a transition state and occurs at
different conductivity values, typically its value is between 10 -11 S cm -1 and 10 -9 S cm -1 . The
relatively high conductivity value of 3.5×10 -7 S cm -1 for sample (B) shows that percolation
threshold has in fact been reached at a significantly lower concentration than 5% PANI.
Sample (C) is a quinary blend containing 5% PANI comprised of
HDPE/PS/PMMA/PVDF/PANI. This system attains an even higher conductivity value of
1.2×10 -6 S cm -1 thus corresponding to a lower percolation threshold due to the further
geometrical restrictions imposed by the additional phase.
Sample (D) represents a blend with five components and a di-block copolymer of adjacent PS
and PMMA phases in the blend. Since the PS and PMMA phases are located beside each other,
the addition of their copolymer to a blend results in the size of the PS and PMMA phases being
decreased (compare Figures 4-13a and 4-13b). This reduction of size in the PS and PMMA
phases carries over to other phases including PANI. The highest conductivity value for blend
containing 5% PANI is obtained for sample (D) with a value of 4.2×10 -6 S cm -1 .
The 5% percolation threshold value for PANI reported above is the lowest ever obtained in a
melt processed multi-component system. Moreover, as discussed earlier, since the PANI sample
used here is a mixture of 25% PANI and 75% zinc compound, the actual concentration of pure
PANI at this percolation threshold value is less than 5%.
Points (E), (F), and (G) represent ternary, quaternary, and five-component blends, each of which
contains 10% PANI. Point (H) is a quaternary blend containing a copolymer of PS and PMMA.
Although the extent of PANI is ten percent in these cases and higher conductivity values are
obtained for blends compared to samples containing 5%, a similar behavior in increasing the
conductivity value with multi-percolated samples is observed. Higher concentration of PANI in
the samples containing 10% PANI results in more connected pathways of PANI in the sample
yielding a higher conductivity range from 6.2×10 -11 S cm -1 for ternary to 2.2×10 -3 S cm -1 for the
quaternary blend plus copolymer.
Ternary, five-component, and six-component samples denoted as (I), (J), and (K), respectively,
having 20% PANI are blended. Sample (I) shows a high conductive value of 2.2×10 -3 S cm -1 for
the ternary blend indicating that although the PANI is encapsulated by the middle
130

phase(PMMA), the concentration of PANI is high enough (20 vol%) to create a random network
inside the PMMA. This implies that at a high concentration of PANI, the effect of the
morphology of the system on the conductivity is less important. At those higher concentrations,
random connected pathways form in the system for all morphologies. But morphology is still
affective whereas samples (J) and (K) comprising of more components represent higher
conductivity values of 1.2×10 -3 S cm -1 and 4.5×10 -3 S cm -1 . Samples L and M in Figure 4-14
with 25% PANI demonstrate that 25% of PANI in the sample is sufficient to completely
diminish the effect of morphology on conductivity and random networks of PANI form
everywhere in the sample. Even at higher concentrations of PANI no significant increase in
conductivity for the blend is observed as random pathways saturate the sample. Since the PANI
concentration and morphology of the phases in these samples have no affect on the conductivity,
even ternary(sample N) and binary(sample O) blends show the maximum conductivity and
plateau behavior.
4.5 Conclusions
We have used polyaniline as a conductive polymer to prepare a solid, 3D, low percolation
threshold conductive device through the control of multiple encapsulation and multiple
percolation effects in a 5 component PE/PS/PMMA/PVDF/PANI polymer blend system through
melt processing. The percolation threshold in the multi-component polymer blend is shown to be
sensitive to the morphological continuity of the various encapsulated phase networks. PANI is
situated in the core of the multiple network system due to its high surface tension and polarity
and, in this way its percolation threshold can be reduced to less than 5%. The detailed
morphology and continuity diagrams of binary, ternary, quaternary and finally quinary systems
are progressively studied in order to systematically demonstrate the concentration regimes
resulting in the formation of these novel multiple-encapsulated morphological structures. First,
an onion-type morphology of HDPE/PS/PMMA/PVDF is prepared which is comprised of an
HDPE matrix with a multiple-component polymeric droplet phase consisting of a PS shell, a
PMMA middle layer and PVDF in the core. The order of phases in this concentrically organized
onion morphology is thermodynamically controlled and is described by Harkins spreading
131

theory. Through the control of the composition of the inner(HDPE) and outer(PVDF) layers in
the onion structure, the morphology can be transformed to a highly elongated onion morphology
and subsequently to a multi-percolated structure in which all HDPE, PS, PMMA, and PVDF
phases are continuous and percolated throughout the system. It is shown that the addition of
PANI results in a 5-component multi-encapsulated, multi-percolated blend with the PANI
situated in the core of the system. The addition of an interfacial modifier for the PS and PMMA
components results in the diminished phase sizes of those particular components which, due to
the encapsulated nature of the morphology, has the effect of reducing the phase size for all other
phases as well. The conductivity is measured for a wide range of the above systems as a function
of the number of phases and as a function of the PANI concentration. It is shown that increasing
the number of components in the blend with a multi-percolated structure from ternary to
quaternary to quinary and finally to an interfacially modified quinary blend results in a several
orders of magnitude increase in the conductivity of the system. In this way, the percolation
threshold of the PANI can be reduced to below 5 vol%. The conductivity of the quinary blend
system, for example, can be increased from 10 -15 S cm -1 (pure PE) to 10 -5 S cm -1 at 5 % PANI
and up to 10 -3 S cm -1 for 10% PANI. These are the highest conductivity values ever reported for
these PANI concentrations in melt processed systems.
4.6 Acknowledgements
The authors express appreciation to the Natural Sciences and Engineering Research Council of
Canada for supporting this work. They also thank F. Normandin from the Institut des matériaux
industriels for his assistance with the conductivity measurements of the samples.
4.1. References
(1) Moussaif, N.; Jérôme, R. Polymer 1999, 40, 3919.
(2) Reignier, J.; Favis, B. D. Macromolecules 2000, 33, 6998.
(3) Reignier, J.; Favis, B. D. AIChE Journal 2003, 49, 1014.
(4) Zhang, J.; Ravati, S.; Virgilio, N.; Favis, B. D. Macromolecules 2007, 40, 8817.
(5) Virgilio, N.; Marc-Aurele, C.; Favis, B. D. Macromolecules 2009, 42, 3405.
132

(6) Virgilio, N.; Desjardins, P.; L'Esperance, G.; Favis, B. D. Macromolecules 2009, 42,
7518.
(7) Torza, S.; Mason, S. G. J. Coll. Interface Sci. 1970, 33, 67.
(8) Hobbs, S. Y.; Dekkers, M. E. J.; Watkins, V. H. Polymer 1988, 29, 1598.
(9) Reignier, J.; Favis, B. D. Polymer 2003, 44, 5061.
(10) Zilberman, M.; Siegmann, A.; Narkis, M.; 1 ed.; SAMPE: Anaheim, CA, USA, 1998;
Vol. 43, p 528.
(11) Zilberman, M.; Siegmann, A.; Narkis, M. Journal of Macromolecular Science -
Physics 2000, B39, 333.
(12) Pud, A.; Ogurtsov, N.; Korzhenko, A.; Shapoval, G. Progress in Polymer Science
(Oxford) 2003, 28, 1701.
(13) Anand, J.; Palaniappan, S.; Sathyanarayana, D. N. Progress in Polymer Science
(Oxford) 1998, 23, 993.
(14) Bhattacharya, A.; De, A. Journal of Macromolecular Science - Reviews in
Macromolecular Chemistry and Physics 1999, C39, 17.
(15) Scher, H.; Zallen, R. The Journal of Chemical Physics 1970, 53, 3759.
(16) Gubbels, F.; Blacher, S.; Vanlathem, E.; Jerome, R.; Deltour, R.; Brouers, F.; Teyssie,
P. Macromolecules 1995, 28, 1559.
(17) Gubbels, F.; Jerome, R.; Teyssie, P.; Vanlathem, E.; Deltour, R.; Calderone, A.;
Parente, V.; Bredas, J. L. Macromolecules 1994, 27, 1972.
(18) Jagur-Grodzinski, J. Polymers for Advanced Technologies 2002, 13, 615.
(19) Bridge, B.; Tee, H. International Journal of Electronics 1990, 69, 785.
(20) Carmona, F. Physica A: Statistical and Theoretical Physics 1989, 157, 461.
(21) Munson-McGee, S. H. Physical Review B (Condensed Matter) 1991, 43, 3331.
(22) Sumita, M.; Sakata, K.; Asai, S.; Miyasaka, K.; Nakagawa, H. Polymer Bulletin
(Berlin) 1991, 25, 265.
(23) Burroughes, J. H.; Jones, C. A.; Friend, R. H. Nature 1988, 335, 137.
(24) Paul, R. K.; Vijayanathan, V.; Pillai, C. K. S. Synthetic Metals 1999, 104, 189.
(25) Macdiarmid, A. G.; Jin-Chih, C.; Halpern, M.; Wu-Song, H.; Shao-Lin, M.; Somasiri,
L. D.; Wanqun, W.; Yaniger, S. I. In Proceedings of the International Conference on
the Physics and Chemistry of Low-Dimensional Synthetic Metals (ICSM 84), 17-22
June 1984; 1-4 ed. UK, 1985; Vol. 121, p 173.
(26) Andreatta, A.; Cao, Y.; Chiang, J. C.; Heeger, A. J.; Smith, P. Synthetic Metals 1988,
26, 383.
(27) Ruokolainen, J.; Eerikainen, H.; Torkkeli, M.; Serimaa, R.; Jussila, M.; Ikkala, O.
Macromolecules 2000, 33, 9272.
(28) Hartikainen, J.; Lahtinen, M.; Torkkeli, M.; Serimaa, R.; Valkonen, J.; Rissanen, K.;
Ikkala, O. Macromolecules 2001, 34, 7789.
(29) Panipol http://www.azom.com/details.asp?ArticleID=1197 2000.
(30) Segal, E.; Haba, Y.; Narkis, M.; Siegmann, A. Journal of Polymer Science, Part B:
Polymer Physics 2001, 39, 611.
(31) Shacklette, L. W.; Han, C. C.; Luly, M. H. Synthetic Metals 1993, 57, 3532.
(32) Ikkala, O. T.; Laakso, J.; Väkiparta, K.; Virtanen, E.; Ruohonen, H.; Järvinen, H.;
Taka, T.; Passiniemi, P.; Österholm, J. E.; Cao, Y.; Andreatta, A.; Smith, P.; Heeger,
A. J. Synthetic Metals 1995, 69, 97.
133

(33) Narkis, M.; Zilberman, M.; Siegmann, A. Polymers for Advanced Technologies 1997,
8, 525.
(34) Zilberman, M.; Titelman, G. I.; Siegmann, A.; Haba, Y.; Narkis, M.; Alperstein, D.
Journal of Applied Polymer Science 1997, 66, 243.
(35) Woo Jin, B.; Won Ho, J.; Yun Heum, P. Synthetic Metals 2003, 132, 239.
(36) Planes, J.; Wolter, A.; Cheguettine, Y.; Pron, A.; Genoud, F.; Nechtschein, M.
Physical Review B (Condensed Matter) 1998, 58, 7774.
(37) Morgan, H.; Foot, P. J. S.; Brooks, N. W. Journal of Materials Science 2001, 36,
5369.
(38) Qinghua, Z.; Xianhong, W.; Yanhou, G.; Dajun, C.; Xiabin, J. Journal of Polymer
Science, Part B (Polymer Physics) 2002, 40, 2531.
(39) Moon Gyu, H.; Seung Soon, I. Polymer 2001, 42, 7449.
(40) Chipara, M. I.; Munteanu, I.; Notingher, P.; Chipara, M. D.; Reyes Romero, J.;
Negreanu, B.; Panaitescu, D.; IEEE: Vasteras, Sweden, 1998, p 372.
(41) Rodrigues, P. C.; Akcelrud, L. Polymer 2003, 44, 6891.
(42) Dong-Uk, C.; Kitae, S.; Young Chul, K.; Yun Heum, P.; Jun Young, L.; Gordon
&amp; Breach: Kyoto, Japan, 2001; Vol. 370, p 399.
(43) Ong, C. H.; Goh, S. H.; Chan, H. S. O. Polymer Bulletin (Berlin) 1997, 39, 627.
(44) Cao, Y.; Smith, P.; Heeger, A. J. Synthetic Metals 1992, 48, 91.
(45) Yang, C. Y.; Cao, Y.; Smith, P.; Heeger, A. J. Polymer Preprints, Division of Polymer
Chemistry, American Chemical Society 1993, 34, 790.
(46) Privalko, V. P.; Ponomarenko, S. M.; Privalko, E. G.; Lobkov, S. V.; Rekhteta, N. A.;
Pud, A. A.; Bandurenko, A. S.; Shapoval, G. S. Journal of Macromolecular Science -
Physics 2005, B44, 749.
(47) Bliznyuk, V. N.; Baig, A.; Singamaneni, S.; Pud, A. A.; Fatyeyeva, K. Y.; Shapoval,
G. S. Polymer 2005, 46, 11728.
(48) Fraysse, J.; Planes, J.; Dufresne, A.; Taylor and Francis Inc.: Poznan, 2000; Vol. 354,
p 511.
(49) Levon, K.; Margolina, A.; Patashinsky, A. Z. Macromolecules 1993, 26, 4061.
(50) Haba, Y.; Segal, E.; Narkis, M.; Titelman, G. I.; Siegmann, A. Synthetic Metals 2000,
110, 189.
(51) Narkis, M.; Haba, Y.; Segal, E.; Zilberman, M.; Titelman, G. I.; Siegmann, A.
Polymers for Advanced Technologies 2000, 11, 665.
(52) Cox, W. P.; Merz, E. H. Journal of Polymer Science 1958, 28, 619.
(53) Yasuda, K.; Armstrong, R.; Cohen, R. Rheologica Acta 1981, 20, 163.
(54) Virgilio, N.; Favis, B. D.; Pepin, M.-F.; Desjardins, P.; L'Esperance, G.
Macromolecules 2005, 38, 2368.
(55) Wu, S. Polymer Interface and Adhesion; Marcel Dekker Inc.: New York, 1982.
(56) Baradie, B.; Shoichet, M. S. Macromolecules 2003, 36, 2343.
(57) Reignier, J.; Favis, B. D.; Heuzey, M.-C. Polymer 2003, 44, 49.
134

ONION MORPHOLOGY
HDPE
PVDF
PS
Control of
the morphology
Table of Contents
MULTI-PERCOLATED
MORPHOLOGY
Addition of PANI
Solvent extraction
135
POLYANILINE NETWORK

CHAPTER 5 - 3D POROUS POLYMERIC MATERIAL WITH LOW
5.1 Abstract
PERCOLATION THRESHOLD OF CONDUCTIVE POLYMER
PREPARED USING LBL DEPOSITION
In this work a novel 3D porous polymeric conducting device (PPCD) is derived from multipercolated
polymer blend systems. The work has focused on the preparation of ultra-low surface
area porous substrates followed by the deposition of polyanilene conductive polymer (PANI) on
the internal porous surface using a layer-by-layer technique. The approach reported here allows
for the percolation threshold concentration of polyaniline conductive polymer (PANI) to be
reduced to values as low as 0.19%. Furthermore, depending on the amount of PANI deposited,
the conductivity of the porous substrate can be controlled from 10 -15 S cm -1 to 10 -3 S cm -1 .
Ternary and quaternary multi-percolated systems comprised of high-density
polyethylene(HDPE), polystyrene(PS), poly(methyl methacrylate)(PMMA) and poly(vinylidene
fluoride)(PVDF) are prepared by melt mixing and subsequently annealed in order to obtain large
interconnected phases. Selective extraction of PS, PMMA and PVDF result in a fully
interconnected porous HDPE substrate of ultra low surface area and highly uniform sized
channels. This provides an ideal substrate for subsequent polyaniline(PANI) addition. Using a
layer-by-layer(LbL) approach, alternating poly(styrene sulfonate)(PSS)/PANI layers are
deposited on the internal surface of the 3-dimensional porous polymer substrate. The PANI and
sodium poly(styrene sulfonate)(NaPSS) both adopt an inter-diffused network conformation on
the surface. The sequential deposition of PSS and PANI has been studied in detail and the mass
deposition profile demonstrates oscillatory behavior following a zigzag-type pattern. The
presence of salt in the deposition solution results in a more uniform deposition and more thickly
deposited PSS/PANI layers. The conductivity of these samples was measured and the
conductivity can be controlled from 10 -15 S cm -1 to 10 -5 S cm -1 depending on the number of
deposited layers. Applying a load to the substrate can be used as an additional control parameter.
Higher loads result in higher conductivity values with values as high as 10 -3 S cm -1 obtained. The
136

work described above has focused on very low surface area porous substrates in order to
determine the lowest possible percolation threshold values of polyaniline, but high surface area
substrates can also be readily prepared using this approach.
5.2 Introduction
One of the main contributions in polymer physics over the last twenty years has been the
development of electronic devices consisting of conducting polymers(Contractor et al., 1994;
Heywang & Jonas, 1992; Kaneto, Kaneko, Min, & MacDiarmid, 1995; Kraft, Grimsdale, &
Holmes, 1998; Roman, Andersson, Yohannes, & Inganás, 1997; Schmidt, Tegtmeyer, &
Heitbaum, 1995; Wessling, 1994). This work has led to: the thin film deposition and
microstructuring of conducting materials(Heywang & Jonas, 1992); materials for energy
technologies(Roman, et al., 1997); electroluminescent and electrochromic devices(Kraft, et al.,
1998); membranes and ion exchangers(Schmidt, et al., 1995); corrosion protection(Wessling,
1994); sensors(Contractor, et al., 1994) and artificial muscles(Kaneto, et al., 1995). Electron
conductive polymers, which are the result of extended π-conjugation along the polymer
backbone, fall into the class of conductive materials exhibiting semi-conducting behaviour. The
discovery of polymer light emitting diodes(Burroughes et al., 1990), in particular, has brought
considerable attention to the efficacy and lifetime of semiconducting polymer-based electronic
devices. The magnitude of the electrical resistivity, or conductivity, determines the application
field of the device and these polymeric optoelectronic devices can be classified in a number of
different categories including: antistatic applications with a range of 10 -14 -10 -9 S cm -1 ;
electrostatic dissipation applications with a range of 10 -9 -10 -5 S cm -1 ; and semiconducting
applications with a range of 10 -6 -10 0 S cm -1 (Margolis, 1989). By controlling the range of the
conductivity, the development of devices such as polymeric photovoltaic devices(Halls et al.,
1995), polymer-based lasers(Tessler, Denton, & Friend, 1996), and transistors(Garnier, Hajlaoui,
Yassar, & Srivastava, 1994) have received considerable attention recently.
Some of the principal approaches used in the preparation of polymeric conductive devices are the
fabrication of ultrathin films by various strategies such as the Langmuir-Blodgett (LB)
137

technique(Zasadzinski, et al., 1994) and the self-assembled monolayer (SAM) method(Maoz,
Frydman, Cohen, & Sagiv, 2000). As well, layer-by-layer(LbL) assembly allows for a high
degree of control of material properties and architecture at the nanometer scale(Hammond, 2004;
Love, Estroff, Kriebel, Nuzzo, & Whitesides, 2005). In less than 20 years since the introduction
of the LbL technique(Decher, et al., 1992a), the electrochemical study of LbL films has grown
from theory to functional devices, but, in most cases, thin films of conductive polymer are
applied on 2D surfaces. One of the main requirements in this field is to further expand the range
and potential of 3D conductive polymeric devices.
In an A/B heterophasic polymer blend, a low concentration of phase A forms droplets in a matrix
of phase B. As the concentration is gradually increased the percolation point is reached and the
continuity of phase A also increases. Near the phase inversion region a co-continuous
morphology is observed(Paul & Barlow, 1980a). A co-continuous morphology is defined as a
state where each phase is completely continuous throughout the material. The co-continuous
phase structure consists of two fully-interconnected phases which mutually interpenetrate each
other(Bourry & Favis, 1998b). The percolation threshold is defined as the formation of longrange
connectivity in random systems. Mono-disperse olid droplets typically demonstrate a
percolation threshold at values of 0.16 by assuming that the occupation of a site or bond is
completely random. A topic of significant interest in heterophasic polymer blends has been to
examine the various factors which resulting in significant reductions in the percolation threshold
concentration(Gubbels, et al., 1994; Wang & Rubner, 2002). One of the most interesting
approaches towards low percolation thresholds is the development of double percolated
structures(Levon, et al., 2002; Sumita, et al., 1992). Zhang et al(Zhang, et al., 2007) reported on
a thermodynamically driven double percolated morphology in an HDPE/PS/PMMA blend which
was essentially a fully continuous system, PS, encapsulating an already continuous system,
PMMA. This type of structure allowed the percolation threshold for the encapsulating polymer to
be reduced to well below 3%. This work demonstrated the potential of layering one phase onto
another continuous structure as a powerful technique to reducing percolation thresholds.
In a ternary liquid system complete wetting can be described when one phase tends to
spontaneously spread on the second phase in the matrix of third phase. For this purpose, Torza
and Mason(Torza & Mason, 1970) and then Hobbs et al.(Hobbs, et al., 1988) employed modified
Harkins spreading coefficients (λij) which is a simple and useful mathematical expression and
138

thermodynamic explanation to predict the wetting characteristic(phase morphology) of ternary
systems.
Equation 5-1 λ = γ − γ − γ
ij jk ik ij
λij is defined as the spreading coefficient giving the tendency of component (i) to encapsulate or
spread onto component (j) in the matrix of component (k). It can also physically be defined as
the transition between the non-wet and wet states. γij, γik and γjk are the interfacial tensions of the
different polymer pairs. A positive value for the spreading coefficient, such as λij, determines that
phase (i) spreads over phase (j) while negative values for all possible spreading coefficients
indicates separately dispersed phases in a continuous matrix.
In co-continuous polymer blends, the solvent extraction of one phase is a route towards porous
materials with a fully interconnected porosity. A number of studies have shown that the phase
size of co-continuous structures can be closely controlled from about 100 nm to hundreds of
microns. The interfacial tension and an annealing step are critical parameters in this regard. One
of the most important ways to coarsen, or increase, the phase size of a co-continuous network is
melt annealing(Yuan, 2005). It has been shown that the annealing of a PS/PE system could
increase the phase size from 0.9 to 72 μm(Sarazin & Favis, 2003). Yuan et al(Yuan, 2005)
observed a linear time dependence for the coarsening of immiscible co-continuous blends and
proposed a capillary pressure effect as the driving force of the coarsening process during static
annealing. Clearly, highly controlled co-continuous morphologies can be converted into highly
controlled porous materials through the selective extraction of one of the phases.
The layer-by-layer deposition technique to produce a polyelectrolyte multilayer on the surface of
a flat substrate was proposed by Decher et al.(Decher & Hong, 1991b) In the LbL approach, the
adsorption process involves consecutive and alternate deposition of positively and negatively
charged polyelectrolytes driven by electrostatic forces followed by a rinsing step with water. A
number of factors can influence mass deposition by LbL such as: ionic strength(Clark,
Montague, & Hammond, 1997), pH of solution(Shiratori & Rubner, 2000), molecular weight of
polyelectrolyte(Sui, Salloum, & Schlenoff, 2003), concentration of polyelectrolytes(Ferreira &
139

Rubner, 1995), and charge density(Schoeler, Kumaraswamy, & Caruso, 2002). Repetitive
deposition steps provide a precise control over the total thickness of the layers in the range from
a few angstroms up to a few micrometers. The thickness increment after each deposition is
referred to as a growth rate which is dictated by polyelectrolyte geometry, surface charges, and
solution parameters(Decher, 1997). Some of the most widely used polyelectrolytes are sodium
poly(styrene sulfonate), poly(diallyldimethyl-ammonium) chloride, poly(ethyleneimine),
poly(allylamine), poly(vinyl sulfate), and poly(acrylic acid). Recent advances in LBL techniques,
in the non-conductive area, have demonstrated the possibility of templating multilayers onto 3D
scale substrates. Caruso et al. (Caruso, Caruso, & Mohwald, 1998) deposited LBL films onto a
colloidal core. Subsequent removal of the core resulted in a thin-film shell. Roy et al. (Roy, et
al., 2006) deposited LBL films onto a fully interconnected porous PLLA surface and the
subsequent removal of PLLA resulted in a 3D object comprised of a vast nanosheath network of,
high surface area and the highest void volume ever reported for a polymeric substrate.
Rubner et al, (Cheung, et al., 1997; Fou & Rubner, 2002) were the first to apply the LbL
technique in the field of electronically conductive polymers to construct PSS/PANI bilayers onto
thin films in order to prepare a conductive device. Important advances have been made in this
area by employing various conjugated polymers such as polyaniline and polypyrrole(Kim et al.,
2002; Liang, Cabarcos, Allara, & Wang, 2004; Ram, Salerno, Adami, Faraci, & Nicolini, 1999).
Ferreira et al.(Ferreira, et al., 1994) found that the solubility of PANI in an organic media such as
dimethylacetamide(DMA) is much higher than that in water. A solution of doped polyaniline
generally makes it more difficult to achieve the spontaneous adsorption of polyaniline chains
onto a variety hydrophilic and hydrophobic surfaces(Cheung, et al., 1997). The presence of salt
in the solutions has a subtle effect. Generally, it screens the monomer-monomer repulsive
interactions, leading to enhancement of adsorption(Dautzenberg et al., 1994). Salt can play
different roles in polyelectrolyte multilayer(PEM) formation and function, such as controlling the
thickness increment of polyelectrolytes, the permeability(Harris & Bruening, 2000), and the
stability(Dubas & Schlenoff, 2001b) of the multilayer. Most of the work published so far has
revealed an increase in conductivity with an increase of the number of adsorbed PSS/PANI
bilayers by measurement of electrical conductivity of LbL films deposited on a flat
surface(Braga, et al., 2008; Paloheimo, et al., 1995). In most cases conductivity measured by
140

four-point probe increases continuously with addition of layers until a saturation plateau is
reached around the 13 th bilayer or 25 th layer.
The principal objective of this work is to prepare a novel 3D porous polymeric conductive
device, prepared via the LbL deposition of a conductive polyelectrolyte onto a 3D substrate of
fully interconnected porosity generated from a co-continuous polymer blend. The work will
involve the preparation of the blends and control of the phase size. Annealing of the cocontinuous
structure, followed by extraction of the porogen phases will be examined with a view
to generating an ultra-low surface area, yet fully interconnected, porous material. Layer-by-layer
adsorption of PSS/PANI onto the internal surface of the porous device will be carried out and
details such as the growth rate and rate of adsorption and desorption of the polyelectrolytes will
be examined. Finally the conductivity will be measured in order to examine the potential of the
developed devices.
5.3 Experimental Methods
5.3.1 Materials
Commercial homopolymers were used in this study to prepare the conductive substrate.
Poly(methyl-methacrylate) powder of Mw=12000 was obtained from Aldrich. High-density
polyethylene, HDPE with a molecular weight of Mw=79000 was obtained from Petromont and
polystyrene, PS with Mw=290000 was purchased from Dow Chemical. Polyvinylidene fluoride
was obtained from Arkema. The polyelectrolytes used were polyaniline with Mw=20000 supplied
by Aldrich as a polycation and poly(sodium 4-styrenesulfonate), Mw=70000 purchased from
Sigma-Aldrich as polyanion. Densities in the melt states were determined via a fully automated
PVT(pressure-volume-temperature) equipment supplied by Thermoelectron. Volume variation at
controlled temperatures and pressures were determined. Some of the characteristic properties of
the homopolymers are reported in Table 5-1.
Table 5-1. Characteristic Properties of Homopolymers
141

Material
Poly(methyl
methacrylate)
Mw * 10 -3
(g/mol)
Melt Index
(g/10min)
η* × 10 -3 at
25 rad s -1 at
190°C (Pa.s)
Density
(g/cm 3 )
at 20 °C
Density
(g/cm 3 ) at
200 °C
12 - 0.04 a 1.19 1
Polystyrene 290 - 1.5 1.04 0.97
High density
Polyethylene
79 8.1 0.72 0.98 0.85
Polyaniline 20 - - - -
Poly(sodium 4styrenesulfonate)
70 - - 1.158 -
Polyvinyl difluoride - 1.5 1.4 b 1.77 1.6
a reported by Reignier et al.(Reignier & Favis, 2000)
b zero shear viscosity of PVDF is 902,000 Pa.s at 190°C as calculated by the Carreau-Yasuda
model
Epofix resin containing 80-100% of Bisphenol-A-diglycidylether with a viscosity of 50cP at
20°C and Epofix hardener containing triethylenetetramine supplied by Struers company were
employed to fill the pores of the template after LbL deposition. Sodium chloride 99.5% was
purchased from Fluka. Ultrapure water with a resistivity of 18.2MΩ.cm was used in all
experiments and was obtained from a Milli-Q plus system Millipore Biocel. For filtration of the
PANI solution, glass fiber filters were purchased from millipore with a diameter size of 47mm.
5.3.2 Rheological Analysis
A constant stress rheometer (SR 5000, Rheometric Scientific) equipped with a 25 mm parallel
disk geometry was utilized to measure the linear viscoelastic properties of homopolymers
containing 0.2% antioxidant. Measurements were carried out using a parallel-plate geometry
with a gap of about 1.2 mm. The experiments were carried out under a nitrogen atmosphere at a
temperature of 190°C over a frequency range from 0.005 to 79.5 rad s -1 . The disc shaped samples
142

in the rheology tests were compression moulded at 190°C. In order to determine the linear
viscoelastic region, a stress sweep test was performed. The Carreau-Yasuda model(Yasuda, et
al., 1981) was employed to determine the zero shear viscosity of PVDF.
5.3.3 Substrate Preparation
Multi-component blends were prepared via melt-blending under a flow of dry nitrogen in a
30mL Plasti-Corder internal mixer (Brabender) operating at 200°C and 50 rpm. The maximum
shear rate at this speed is close to 25 s -1 . The real temperature achieved by the end of mixing
experiment was approximately 185°C. After converting the mass to volume (densities presented
in Table 5-1), the mixing chamber was filled to 70% of its total volume. All the blends in this
study were mixed for 8 min which is more than sufficient to allow the torque to achieve a
constant value. Irganox antioxidant, supplied by CIBA, was added to the mixture at a level of 0.2
wt% to reduce the thermal oxidation of the components. After mixing, the samples were
immediately cut from the mass and quenched in a cold water bath to freeze-in the morphology.
All concentrations have been reported based on the component volume fraction shown in Table
5-2.
Table 5-2. Composition of Polymer Components in Ternary and Quaternary Blends
Samples HDPE PS PMMA PVDF
A 33.3% 33.3% 0 33.3%
B 33.3% 0 33.3% 33.3%
C 40% 30% 10% 20%
D 40% 10% 40% 10%
143

5.3.4 Annealing Test
For the annealing experiments a compression molding press was employed. Small pieces of
samples were cut from the blend and then sandwiched between two metal plates in a disc-shaped
mould. The temperature of both upper and lower plates was set at 200°C. The annealing was
carried out under nitrogen in five steps. As a first step, both heating plates of the compression
press were brought into contact with the sample pieces without imposing any pressure for 5
minutes, in order to minimize any deformation or flow of the sample. After heating and melting
the samples, pressure was applied for 15 minutes. After annealing, the samples were cooled by
placing the samples in a large cold press for 3 minutes.
5.3.5 Solvent Extraction
Solvent extraction was used to extract various components and generate porosity in the substrate.
Selective solvent extraction also allows for the determination of the % continuity of the different
components in the blend system using Equation 5-2. The volume of the components before and
after extraction is measured by weighing the sample and converting the weight to volume. As a
primary advantage, solvent extraction is an absolute measurement of continuity.
Equation 5-2
% Cocontinui ty
∑V − ∑V
initial
n n
=
∑V
A homogenous blend is assumed when carrying out the selective solvent extraction of the
PMMA, PS, and PVDF phases in dimethylformamide(DMF). Volume loss measurements were
carried out for one week at room temperature and were used to calculate the extent of continuity
of the PMMA, PS, and PVDF phases altogether. Vinitial and Vfinal are the volume of one
component or several components present in the sample before and after extraction calculated by
weighing the sample and converting it to the volume. The % continuity represents the fraction of
a phase that is continuous.
n
initial
final
144

5.3.6 Layer-by-Layer Deposition
The solid substrate was alternatively immersed for 25min in PSS and PANI solutions with
intermediate water rinsing. An aqueous PSS solution of 10 mg/mL buffered to a pH of 2.5 was
used as the negative polyelectrolyte solution. The PANI solution was prepared using a procedure
by Cheung et al.(Cheung, et al., 1997). A water bath with a pH adjusted to 2.5 was used to
remove excess PSS and PANI on the surface of substrate after each dipping. The process was
periodically interrupted and samples were dried first under a gentle flow air and then by putting
samples in the vacuum oven for 30min prior to weighing.
5.3.7 Filling the Porous Sample with a Resin
A two-component epoxy(Epofix) was used to impregnate the pores of the porous substrates.
Epofix resin and Epofix hardener were mixed and stirred carefully for at least 2 minutes by
maintaining a mixing ratio of 15 to 2 parts by volume. The mixture was carefully poured into the
mould. To eliminate air bubbling during the hardening, the Epovac vacuum impregnation unit
was operated for 24h. After evacuating the specimen chamber, the impregnating fluid is then
introduced into the chamber under atmospheric pressure by means of a disposable plastic tube.
5.3.8 Characterization and Phase Morphology
5.3.8.1 Microtomy and Scanning Electron Microscopy
Samples were cut and microtomed using a Leica-Jung RM 2065 equipped with a glass knife and
a cryochamber type (LN 21). Other samples were fractured under liquid nitrogen known as
cryofracture method. After chemical extraction with dimethylformamide to remove all
components except HDPE, a JEOL JSM 840 scanning electron microscope was employed to
observe the morphology of the samples. Photomicrographs were taken at a range of 5-12keV
after coating the surface of the sample with a gold/palladium alloy by plasma deposition.
145

5.3.8.2 Focused Ion Beam and Atomic Force Microscopy
Cryomicrotoming of the specimens using a glass knife was carried out in order to obtain a
perfectly flat surface for subsequent morphology observation. The samples were coated with a
gold-palladium alloy. The surface of the samples was then treated and etched using a Hitachi
focused ion beam FIB-2000A operated at 30 keV gallium beam. Further details of the FIB
preparation are given in Virgilio et al(Virgilio, et al., 2005a).
Finally, the topography of the sample surface was analyzed by tapping mode AFM using a
Dimension 3100 scanning probe microscope equipped with a Nanoscope IIIa control module.
5.3.8.3 Conductivity Measurements
DC electrical measurements were performed through the vertical thickness of the substrate at
ambient temperature using a Keithley electrometer model 6517 working at a voltage range of 0-
1000V and ammeter range of 0-20mA. In-house software was used to convert the data to
conductance. In order to standardize the pressure for all samples, a 5 lb and 10 lb load was
applied on the sample holders. Cylindrical specimens with a thickness of 1mm and a radius of
1.25cm were compression moulded and then electrical tests were performed under dry air.
Although the top and bottom surfaces of the samples are flat, good contact of the sample surface
and electrodes was further ensured by using two graphite sheets which covered the surfaces of
the samples.
5.4 Results and Discussion
5.4.1 Multiple-Percolated Morphology with Large Continuous Domain
Phases
Since one of the principal objectives of this work is to determine the lowest possible percolation
threshold for PANI in a 3D porous polymer conducting device (PPCD), it was necessary to first
146

prepare a polymer blend structure with large fully interconnected phases. Subsequent extraction
of selected phases within that blend would thus allow for the preparation of a very low surface
area, fully interconnected, porous polymer substrate(Figure 5-1). As will be shown below, the
most effective route towards this is via the preparation of a multi-percolated, multi-component
polymer blend.
a) b)
Extraction of phases
Deposition of PSS/PANI layers
Figure 5-1. 3D schematic showing the approach used to prepare the porous polymeric conducting
device via LbL deposition. (a) solid multi-percolated structure comprised of HDPE, PS, PMMA
and PVDF after melt blending and annealing; (b) porous polymeric conducting device after
extraction of PS, PMMA and PVDF, followed by LbL deposition of PSS and PANI.
Recently(J. Zhang, et al., 2007), it has been shown that a ternary co-continuous (double
percolated) morphology could be developed in which hierarchically ordered phases of
HDPE/PS/PMMA were generated with all phases fully continuous. It was shown that this
phenomenon was governed by the Harkins equation and is a thermodynamically driven process.
A positive spreading coefficient of PS over PMMA (λPS/PMMA) results in the PS phase locating at
the intersection of the HDPE and PMMA phases(Guo, Packirisamy, et al., 1997; Reignier &
Favis, 2000). Figure 5-2a illustrates this case where in a 40/20/40 HDPE/PS/PMMA blend,
PMMA phase is encapsulated by PS phase in the matrix of HDPE. By controlling the
concentration of the phases, a region is determined in which all phases are fully continuous. A
40/10/50 HDPE/PS/PMMA ternary blend is another example of this region where the
147

composition of PS was controlled to generate a PS shell down to a few hundred nm between two
continuous PS and PMMA structures (Figure 5-2b). Subsequently, the PS and PMMA phases
can be extracted, yielding a fully interconnected porous HDPE substrate of about 60% void
volume. This method can be employed to prepare substrates with higher void volume. For
instance, in the 40/10/50 HDPE/PS/PMMA ternary blends if one could obtain a selective solvent
for HDPE, the combined removal of HDPE and PMMA would yield a highly porous PS
substrate of about 90% void volume.
a)
b)
HDPE
PMMA
PS
PMMA PMMA
HDPE
Figure 5-2. (a) FIB-AFM topographic surface image of 40/20/40 HDPE/PS/PMMA. The
topographical height vs distance is shown to the right. The white line in the image indicates the
trace line used in the height vs distance analysis. and (b) Scanning electron micrograph of
40/10/50 HDPE/PS/PMMA after extraction of the PS phase by cyclohexane
HDPE
PS
PMMA
148

a)
b)
c)
Figure 5-3. Scanning electron micrographs of binary and ternary samples after extraction of all
phases (except polyethylene) by DMF for and (a) 33/33/33 HDPE/PMMA/PVDF, (b) 33/67
HDPE/PVDF, and (c) 50/50 HDPE/PVDF
149

Figure 5-3 illustrates a comparison between binary co-continuous structures (single percolated)
and a ternary one (double percolated) and shows that continuous phases can be generated at
lower concentrations of each phase in the latter case due to the lower percolation threshold of
those phases.
In Figure 5-3a, it is shown that a double percolated HDPE/PMMA/PVDF blend, after PMMA
and PVDF extraction, can result in a porous substrate of 33% HDPE which demonstrates a
homogeneous, perfectly interconnected structure while a binary blend of 33% HDPE and 67%
PVDF yields an imperfect continuous structure of HDPE (Figure 5-3b). Extraction experiments
indicate that 98% of the PMMA and PVDF were extracted in the ternary blend proving their high
levels of continuity. In the case of the 33%HDPE/67%PVDF binary blend, extraction of PVDF
resulted in a value of 108% of material indicating a significant quantity of HDPE present as
dispersed entities within the PVDF. The fully continuous binary 50% HDPE/50% PVDF blend is
also shown for comparison (Figure 5-3c). Clearly, a double-percolated morphology is a route
towards fully continuous phases at low phase concentrations. The subsequent extraction of
PMMA and PVDF leaves behind a fully interconnected porous HDPE substrate of low surface
area. In this work, as HDPE substrates were exposed to polyelectrolyte solutions, extra-large
pores also facilitated the penetration of the solution inward to the interconnected porous area.
The addition of a fourth component to form a quaternary blend with a triple-percolated
morphology between adjacent phases is also possible providing that the spreading coefficients of
the fourth phase and the engulfed phases satisfy Harkins equations. A more detailed study of the
generation of such structures is presented in a separate paper(Ravati & Favis, 2010). A triplepercolated
morphology is particularly interesting in the present work since it provides a route
towards the generation of porous channels of highly uniform size distribution. Figure 5-4
demonstrates such a structure in which PMMA situates at the interface of PS and PVDF in an
HDPE matrix. Thus this triple-percolated morphology follows the order
HDPE/PS/PMMA/PVDF (Figures 5-4a and 5-4b). Figures 5-4a and 5-4b represent quaternary
blends with the PMMA phase extracted, yielding PS layers engulfed in an HDPE matrix while
the PVDF phase remains at the core encapsulated by extracted PMMA. The removal of the PS
phase separates HDPE/PS/PMMA/PVDF into two parts: the HDPE matrix and PMMA/PVDF
core (Figure 5-4c). These figures unambiguously show that the PS and PMMA phases are
situated between HDPE and PVDF, and a hierarchical structure of four different phases is
150

spontaneously self-assembled. After the removal of all other phases, the final HDPE porous
substrate exhibits a continuous structure with large pore sizes and macro-channels of highly
uniform distribution. The uniformity of the double-percolated morphology represented in Figure
5-3a and that of the triple-percolated one shown in Figure 5-4d are compared morphologically.
The reason for the more uniform structure in the triple-percolated system is the phases are
distributed between three interfaces as opposed to be distributed between two interfaces in the
double-percolated system. Thus, the triple-percolated structure system appears to form phases
which, after extraction, result in a substrate with a more uniform pore size distribution.
a) b)
c) d)
HDPE
PVDF
Figure 5-4. Scanning electron micrographs of quaternary polymer blends for various
compositions compounded via melt-blending (a) 50/16/16/16 HDPE/PS/PMMA/PVDF after
extraction of PMMA by acetic acid, (b) 50/20/10/20 HDPE/PS/PMMA/PVDF after extraction of
PMMA by acetic acid, (b) 50/20/10/20 HDPE/PS/PMMA/PVDF after extraction of PS by
cyclohexane, and d) 40/10/40/10 HDPE/PS/PMMA/PVDF after extraction of all phases except
HDPE
151

Four samples based on the ternary and quaternary systems of HDPE, PS, PMMA and PVDF will
be the principal substrates used in subsequent experiments reported in this paper. These blends
are shown in Table 5-2 and are designated as samples A-33%HDPE/33%PS/33%PVDF; B-
33%HDPE/33%PMMA/33%PVDF; C-33%HDPE/30%PS/10%PMMA/20%PVDF; and D-
33%HDPE/10%PS/40%PMMA/10%PVDF. It is possible to generate porous substrates of an
even higher void volume (lower concentration of HDPE), however the mechanical strength of
those materials is quite low. Since the study of the effect of load on the conductivity of the
sample is also one of the objectives of this work, the above concentrations were selected. These
blends effectively allows for the preparation of an ultra low surface area, fully interconnected
porous substrate of high void volume.
5.4.2 Annealing the Samples in order to Increase Pore Size
It has been shown in previous work that annealing co-continuous polymer blends is an excellent
route towards significantly increasing the phase size and further lowering phase surface area.
During this process of static annealing, thin parts of the continuous network merge into larger
parts via capillary driving forces(Zhenhua Yuan, 2005).
152

a) b)
c) d)
e) f) g) h)
Figure 5-5. Scanning electron microscope images of the morphologies of the porous structures
showing the influence of annealing. The top row is without annealing, the bottom row is after
annealing. All components have been extracted except HDPE. a),e) Sample (A); b),f) Sample
(B); c),g) Sample (C); and d),h) Sample (D).
Double percolated HDPE/PS/PVDF and HDPE/PMMA/PVDF blends and triple-percolated
HDPE/PS/PMMA/PVDF were subjected to quiescent annealing for 15 minutes in a compression
molding press at very low contact pressures. This operation results in a substantial increase in the
average phase size. Figure 5-5 shows the morphology of the blends after annealing and
subsequent extraction of all phases except for HDPE. The phase size increase is dramatic and is
shown for sample B to increase from approximately 40μm to 500μm as exhibited
morphologically in Figures 5-5b and 5-5c. The phase increase before and after annealing for the
other samples A, C, and D is also substantial and ranges from 5 to 15 times (Figure 5-5). The
results indicate that the coalescence of a multi-percolated system occurs much faster than
coalescence in a binary blend. Yuan et al.(Zhenhua Yuan, 2005) reported that, in the best case
scenario, a binary blend requires approximately 40 minutes to coarsen from several microns to a
phase size of about 160 microns. Consequently, the required annealing time to achieve 400-500
microns is estimated at about two hours of annealing. It is proposed that geometrical restrictions
in a multi-percolated system enormously accelerate coalescence effects while still maintaining
full continuity of all components.
153

5.4.3 Continuity and Surface Areas of Substrates
Table 5-3 shows gravimetric data obtained after solvent extraction and demonstrates that in all
cases the various phases within the HDPE are fully continuous. The continuity of all extracted
phases for both annealed and unannealed samples is higher than 96%. Hence, the 33% volume
fraction of HDPE in samples A and B and 40% in samples C and D results in a 67% and 60%
void volume respectively after the extraction of all other phases. As well, the effect of annealing
on reducing the surface area of the porous substrate is extremely evident as it was below the
range which could be measured quantitatively by BET (Table 5-4).
Table 5-3. Continuity of the Porous Morphology With and Without annealing (PS, PMMA and
PVDF extracted).
Sample
Continuity of (PS+PMMA+PVDF)
no annealing
Continuity of (PS+PMMA+PVDF)
with annealing
A 96.6 % 98.3 %
B 98 % 100 %
C 98.3 % 99.3 %
D 96 % 97.4 %
Table 5-4. Surface Area Measurements of Porous Polymer Substrates by BET
Sample Surface area measured by BET(m2 /g)
no annealing
Surface area measured by BET(m 2 /g)
with annealing
A 1.28 < 0.01
B 0.70 < 0.01
C 0.88 < 0.01
D 0.74 < 0.01
154

Since the ultra-low surface area values are beyond the range of the BET equipment, the surface
area of sample A (morphology shown in Figure 5-5e) was estimated arithmetically. A porous
disc-shape sample with a diameter of 2.5-cm and a length of 1-mm is assumed to be comprised
of numerous parallel cylinders from top to bottom with a diameter of 175-µm and a length of 1mm.
The diameter of cylinders is obtained from average diameter of HDPE cylindrical rods
exhibited in Figure 5-5e. Estimating the number of cylinders and the surface area per cylinder as
6818 and 0.00375m 2 /g, respectively, corresponds to a surface area of the porous device of
approximately 0.01 m 2 /g.
Hence, the work to this point has achieved fully interconnected porous substrates (samples A, B,
C, and D) of ultra-low surface area (large pores) and a highly uniform pore size. All these
samples will be used as substrates for the layer-by-layer deposition of polyaniline in the
subsequent parts of this work. Table 5-5 compares the surface area of porous sample A to the
surface area of other co-continuous systems and demonstrates that, to our knowledge, this is the
lowest surface area, fully interconnected porous polymer material ever presented in the literature.
Table 5 also demonstrates the potential of preparing conductive devices from higher surface area
co-continuous systems and underlines the versatility of the approach.
155

Table 5-5. Comparison of Surface Areas Reported in Previous Studies and in the Current Work
(last two rows).
porous samples prepared by melt-blending from this laboratory
50/50 PLLA/PCL after LbL deposition of PSS/PDADMAC and
extraction of both phases(Roy, et al., 2006)
50/50 PLLA/PCL after 2h annealing and after LbL deposition of
PSS/PDADMAC and extraction of both phases(Roy, et al., 2006)
40/60 HDPE/SEB (styrene-ethylene-butylene diblock
copolymer)(Jianming Li, et al., 2002)
50/50 HDPE/SEB (styrene-ethylene-butylene diblock
copolymer)(Jianming Li, et al., 2002)
50/50 HDPE/SEBS(styrene-ethylene-butylene-styrene triblock
copolymer)(J. Li & Favis, 2001a)
156
reported specific
surface area measured
by BET(m 2 /g)
140
63
16
13.15
40/60 HDPE/PS after extraction of PS(J. Li & Favis, 2001a) 0.82
33/33/33 HDPE/PMMA/PVDF after extraction of PMMA and PVDF 0.7
33/33/33 HDPE/PMMA/PVDF after 15min annealing and extraction
of PMMA and PVDF
5.4.4 Layer-by-Layer Deposition of PSS/PANI
8.2
≤0.01
A layer-by-layer deposition technique was used to deposit PSS and PANI alternately on the
internal surface of the prepared porous substrate (Figures 5-6a and 5-6b). A PSS polyelectrolyte
soluble in water was selected to be deposited first due to its ability to adsorb strongly on
hydrophobic surfaces such as HDPE(R. v. Klitzing, et al., 1999). In a subsequent step, a semi-
flexible PANI layer is deposited on a PSS layer as a result of a balance between electrostatic
repulsion of charged monomers. Up to thirty-eight PSS and PANI layers were deposited. In order
to visually examine the multilayer assembly on the surface of the HDPE substrate, the void
volumes or empty macro-channels were refilled with epoxy resin.

c)
a)
b)
Figure 5-6. a),b) Scanning electron micrographs of the morphology of sample (B) after
extraction of PMMA, PVDF and deposition of 38 PSS/PANI layers; c),d) SEM and FIB-AFM of
sample(B) after extraction of PMMA, PVDF and deposition of 38 PSS/PANI layers. Pores are
filled with epoxy resin.
d)
157

Microstructural information collected by SEM (Figure 5-6c) and AFM (Figure 5-6d) depicts a
relatively thick polyelectrolyte multilayer as wide as 5.5μm for 38 layers of PSS and PANI on
the surface of the filled sample (B). Picart et al.(Picart, Lavalle, et al., 2001) also reported a very
thick multilayer film of 1μm for 16 layers. They proposed that this thickness increase can be
attributed to a diffusion of polyelectrolyte inwards and outwards in the multilayer during the LbL
process. They also reported an exponential growth of layer thickness similar to the mechanism of
mass growth in this work.
5.4.5 Mass Deposition Dynamics of PSS/PANI
Although the structural information for the construction of ultra-thin multi-layers is extremely
difficult to obtain, gravimetric measurements can interpret the molecular topology of the films
that play the most important role in the formation of multi-layers. Figure 5-7 shows the
quantification of mass deposition of each layer as measured by gravimetric techniques for
various substrates in both the absence and presence of salt. It shows an unusual oscillating
behavior of thickness growth. After dipping the substrate in the PSS solution, the deposition
mass increases, followed by a decrease in deposition mass in a PANI solution due to the partial
removal of a previously adsorbed layer. Some previous works(Dubas & Schlenoff, 2001c;
Schoeler, et al., 2002; Tjipto, Quinn, & Caruso, 2007) have also reported on the oscillating
behavior of PSS polyelectrolytes due to the partial removal of previously adsorbed layers in
subsequent operations. Schoeler et al.(Schoeler, et al., 2002) related the extent of removed
material to the critical charge density of the polyelectrolyte. Charge density for a polyelectrolyte
is defined as the fraction of monomers in the polymer which are charged. If this fraction is much
smaller than 1, the polyelectrolyte is weakly charged. Normally, weak polyelectrolytes tend to
demonstrate an oscillating deposition. Dubas et al.(Dubas & Schlenoff, 2001c) concluded that
the drawing out of water molecules unassociated with specific ion pairs due to increasing
external osmotic pressure is the main reason for the decrease in thickness at low salt
concentrations. Tjipto et al.(Tjipto, et al., 2007) related the oscillating behavior of a multilayer
formed of poly(4-styrenesulfonic acid-co-maleic acid) and PDADMAC to the heterogeneity and
roughness of the samples. Later in this paper it will be shown that all of these explanations are
likely responsible for the oscillating deposition observed in this work. The addition of 1 molar
158

salt to a PSS polyelectrolyte solution magnifies the extent of the oscillation behavior (Figures 5-
7a-d) and this will also be discussed in more detail in the next section.
a)
b)
Mass Increase(%)
Mass Increase(%)
1,8
1,6
1,4
1,2
1
0,8
0,6
0,4
0,2
0
2,2
2
1,8
1,6
1,4
1,2
1
0,8
0,6
0,4
0,2
0
PANI
PSS
Sample (A) with salt
0 5 10 15 20 25 30 35
Number of layers
PANI
PSS
Sample (B) with salt
Sample (A) without salt
Sample (B) without salt
0 5 10 15 20 25 30 35
Number of layers
159

c)
d)
Mass Increase(%)
Mass Increase(%)
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
1,4
1,2
1
0,8
0,6
0,4
0,2
0
PANI
PSS
Sample (C) with salt
0 5 10 15 20 25 30 35
Number of layers
PANI
PSS
Sample (D) with salt
Sample (C) without salt
Sample (D) without salt
0 5 10 15 20 25 30 35
Number of Layers
Figure 5-7. Mass increase (%) indicating cumulative deposited mass as a function of number of
deposited PSS and PANI layers for solutions containing salt and without salt. a) sample(A); b)
sample (B); c) sample(C); and d) sample (D).
160

A relationship between macropore sizes and mass deposition of layers is found. Approximate
average macropore sizes for samples A, B, C and D after annealing are 300μm, 500μm, 100μm,
and 150μm respectively (see Figure 5-5). Figure 5-7 represents the maximum mass deposition
for each sample: sample A attains 1.6% after deposition of 31 PSS/PANI layers and 0.82% after
deposition of 32 layers. Sample (B) attains 2% and 1.16% after deposition of 31 and 32 layers.
Samples C and D attain about 0.8% for 31 layers and 0.5% for 32 layers. These results confirm
that larger pore sizes facilitates the penetration of solution and thicker layers constructs
5.1.1. Diffuse Network Structure of PSS/PANI
Lavalle et al.(Lavalle, et al., 2002), using molecular labeling, observed a diffusion of a portion of
polyelectrolytes in and out of the film in a nonlinear manner leading to exponential growth. This
was observed when one of the polyelectrolytes used was weak as in the current work with PANI.
Schlenoff et al.(Dubas & Schlenoff, 2001b; Schlenoff & Dubas, 2001) have reported that the
surface charge, propagated and inverted by sequential adsorption steps, is overneutralized
(overcompensated) at each step. In that work, the multilayer/solution interface, the charge of the
last-added polymer compensates the previous one by a factor φ. The opposite polymer charges
are in exact stoichiometric ratio for φ=1 yielding a steady-state thickness increment vs layer
number. The overcompensation factor, phi, can be defined for multilayers when the charge factor
has a value higher than 1. Overcompensation leads to excess charge density from the last-added
polyelectrolyte, but only at the surface is φ “unrestricted” by bulk effects. At increasing length, l,
from the interface into the multilayer, the level of overcompensation decreases in an exponential
fashion demonstrating an overcompensation gradient(Schlenoff & Dubas, 2001).
The diffusion of a portion of polyelectrolytes in and out of the film in a nonlinear manner leading
to exponential growth was proposed as a model for overcompensation(Lavalle, et al., 2002). In
this mechanism, the first population was constituted of chains directly interacting with the
surface of the multilayer, which are responsible for linearly growing films. The second
population was constituted of free chains that diffuse into the multilayer over its whole surface
and are responsible for exponential growth. In this study, when a substrate is brought into contact
with a charged PSS solution, many free PSS chains paired with hydrated counter-ions in the
solution diffuse throughout the film in the x, y, and z directions due to electrostatic interactions.
161

Following this, the desorption or detachment of some layers is expected in an oppositely charged
PANI solution when mobile chains diffuse out. As soon as they reach the surface, an additional
reaction occurs between them and the oppositely charged chains available in the solution and an
extra layer is formed. In this fashion, a diffuse network structure throughout the deposited
thickness is observed in contrast to the formation of discrete polyelectrolyte layers in classic
layer-by-layer deposition where strong polyelectrolytes are used.
Based on the above discussion, a schematic representation of film growth has been made to show
the progressive formation of multilayers (Figure 5-8). This schematic contrasts with the classic
formation of well-separated hierarchical alternating multilayers observed in most work using the
LbL approach. In the current work, the oscillating deposition indicates that PSS and PANI
polyelectrolytes diffuse into the previously deposited layers. Like chains make contact with other
like chains and consequently, an inter-diffused network of PANI and PSS is generated. In a
following section, it will be shown that an increase in the conductivity is observed as a function
of the number of PSS and PANI layers and it further confirms the formation of a continuous
network of PANI interdiffused with PSS. The degree of interconnection of this network becomes
more pronounced as the number of deposited layers increases. As well, an increase in the number
of deposited layers results in a rougher surface, leading to a further diffusion of free chains.
Figure 5-8. Schematic of exponential growth of PSS/PANI multilayer thickness on the internal
surface of the substrate.
5.4.6 The Effect of Salt on the Mass Deposition Growth of PSS/PANI
Salt is generally added in procedures involving LbL deposition. In the absence of salt, charged
PSS and PANI in the solution adopt a more expanded rigid, rod-like configuration due to
coulomb repulsion(Ha & Thirumalai, 1995). The role of salt is to gradually screen the monomermonomer
repulsive interactions between PSS charges with counter-ions and also bring them
162

close together. It is interesting to note in Figure 5-7 that the addition of salt significantly
increases the oscillation behavior of the mass growth dynamics.
This oscillation behavior can be explained as follows: the number of PSS molecules that diffuse
throughout the multilayer and the total concentration of these free chains depend on the
electrostatic interaction forces that arise from total overcompensation. Two important parameters
influencing the extent of overcompensation are the concentration of added salt and the number of
layers. Thus, extra overcompensation is caused by the presence of NaCl in the PSS
solution(Klitzing & Moehwald, 1995). As well, the displacement of hydrated small salt counterions
and charged polymer segments result in highly swollen polymers as attached water
molecules are brought in(Dubas & Schlenoff, 2001c). Jaber et al. (Jaber & Schlenoff, 2007)
showed that salt counter-ions can thermodynamically control up to 40% of the mass deposition
of the multilayer construction due to this additional water content. Swelling and
overcompensation occur simultaneously during the construction of the multilayer in the presence
of salt(Dubas & Schlenoff, 2001b), which is involved in the dramatic mass increase of the
multilayer.
In a subsequent dipping step in the PANI solution, the electrostatic interaction between the
diffused hydrated PSS with PANI chains inside the solution play a major role in the diffusion of
free PSS chains out of the multilayer. At this step, in the presence of salt, a significant decrease
in mass deposition is detected which is proportional to the mass increase in the previous
step(PSS adsorption) as compared to a slight decrease in mass in solutions without salt.
Consequently, the amount of thickness growth highly depends on the amount of salt that affects
the electrostatic interactions and increases the driving force for diffusion in and out of the
multilayer.
Mass deposition growth for the PSS/PANI system in cases where salt is both present and absent,
is not linear since a progressive increase of each polyelectrolyte after its deposition is detected
(Figure 5-9). As the number of layers increases, the extent of free chains going inward and
outward increases exponentially due to increase the amount of overcompensation (Figure 5-9). In
other words, at each adsorption step, more PSS is adsorbed than that removed in the previous
step, leading to nonlinear layer growth. The rates of increase and decrease of PSS and PANI at
163

each step corresponding to diffusion in and out of the multilayer is almost double for solutions
containing salt (Figure 5-9).
a)
Mass Increase(%)
b)
Mass Increase(%)
0,6
0,5
0,4
0,3
0,2
0,1
0
0,25
0,2
0,15
0,1
0,05
0
PANI
PSS
0.0767 0.1432
3 4 5 6 7 8 9 10 11 12
PANI
0.0283 0.0544
0.1068
0.1955
0.0304 0.0848
0.159
0.2364
Number of layers
0.0522 0.071
3 4 5 6 7 8 9 10 11 12
Number of layers
0.2045 0.2955
0.0478 0.1196
164

Figure 5-9. Mass increase (%) indicating cumulative deposited mass as a function of number of
deposited PSS and PANI layers for sample (A). a) solution containing salt, and b) solution
without salt.
5.4.7 Conductivity Measurement of the Conductive Porous Polymer Device
The electrical conductivity of sample A-33%HDPE/33%PS/33%PVDF and sample B-
33%HDPE/33%PMMA/33%PVDF was evaluated as a function of the number of deposited PSS
and PANI layers. The electrical conductivity of these samples increases by several orders of
magnitude as a function of the number of deposited PSS/PANI layers until a saturation plateau is
reached at approximately 32 layers (Figure 5-10).
Conductivity(S/cm)
1,0E-02
1,0E-03
1,0E-04
1,0E-05
1,0E-06
1,0E-07
1,0E-08
1,0E-09
1,0E-10
1,0E-11
1,0E-12
1,0E-13
1,0E-14
1,0E-15
0 8 16 24 32 40
Number of layers
Figure 5-10. Conductivity as a function of the number of deposited layers for various samples at
a 5 lb load. The effect of a 10 lb load is shown for two samples as indicated by the arrows.
165
Sample (A) (5lb)
Sample (B) (5lb)
Sample (C) (5lb)
Sample (D) (5lb)
Sample (B) (10lb)
Sample (C) (10lb)
Sample (B) (5lb)
without salt

The percolation threshold of PANI in this porous device is achieved at a maximum of 8 layers.
This corresponds to 0.19 wt.% PANI for sample A and 0.28 wt.% for sample B. The
conductivity saturation plateau is achieved at 32 layers which corresponds to 0.76% PANI for
sample A and 1.1% for sample B. This dependence of conductivity on deposited layers confirms
the hypothesis that deposited PSS and PANI form a network type construction as opposed to the
formation of discrete PSS and PANI layers. In fact, this network formation allows for the fine
tuning of conductivity over several orders of magnitude. Clearly, the deposition of only a few
layers results in an incomplete network with disconnections in many parts and demonstrates a
limited passage of electricity and thus yields a low conductivity value. By increasing the number
of layers to 8 and 16, higher conductivity values are obtained and the conductivity is in the 10 -9 -
10 -8 S cm -1 range. As more PANI chains are added they diffuse into the network resulting in a
significant increase in network branches. The addition of further layers results in a plateau value
for conductivity at approximately 10 -6 S cm -1 for sample A and 10 -5 S cm -1 for sample B. Beyond
32 layers, the further addition of PANI has no effect on the conductivity of the device. The
conductivity of sample C and D after the deposition of 38 PSS/PANI layers was measured
showing the values of 10 -9 S cm -1 and 3×10 -7 S cm -1 respectively. The difference in the
conductivity values can be dependent on several factors such as pore sizes and pore distribution;
however the most important one is the internal surface area of the substrate.
The addition of salt during the LbL process has little influence on the conductivity of the sample
as shown in Figure 5-10. Braga et al. (Braga, et al., 2008) and Paloheimo et al.(Paloheimo, et al.,
1995) reported that the electrical resistance of PANI/PSS layers with solutions of different
polymer concentrations decreases as more layers are adsorbed until a saturation plateau is
reached, between the 26 th and 30 th layer. This closely corresponds with the results observed in
this work. It is very likely that the formation of a diffuse network structure in the current study is
critical in obtaining high conductivity values. Discrete molecular layers added in a classic LbL
protocol using strong polyelectrolytes would have likely resulted in a lower percolation threshold
value, but also would not be capable of achieving the high conductivities observed at saturation
as seen in this study.
166

In this work the conductance of the porous samples was converted to conductivity based on the
general following equation:
Equation 5-3.
Where G is conductance, A is cross-sectional area, l is length, and σ represents conductivity. As
void domains and HDPE have conductivity values of zero and 10 -15 S cm -1 , respectively, a
sample including substrates, multilayers and void volumes can be considered as a whole
conductive device with certain conductivity. In other words, a sample can be considered as an
ultra-porous PANI network. In porous materials, pore geometry (including the pore size and
distribution) plays a major role in the conductivity of the material(Solonin & Chernyshev, 1975),
as well as the perfection of the contact between the conductive material and the nature of the
material filling the pores(Montes, Cuevas, Rodriguez, & Herrera, 2005). A number of theoretical
and experimental models have been developed for applying porous-structure parameters, such as
volume and pore concentration to calculate conductivity(Lifshitz, Landau, & Pitaevskii, 1984;
Skorokhov, 1972). The detailed calculations for the estimation of conductivity using this
approach for this type of porous device is described in more detail elsewhere(Ravati, 2010).
5.4.8 Conductivity of the Conductive Porous Polymer Device under Loads
The conductive porous polymer device developed in this study is sensitive to applied
compression providing that the void volume percentage is sufficiently high. The compression of
sample B of 67 void % under a 10lb load pushes and forces the rods and walls of the substrate to
move inward. On the other hand, no deformation in sample C, which has a 60% void volume, is
observed. In that case the mechanical strength of the walls of the device is sufficient to resist
deformation resulting from the applied load. In the case of sample B, further contact between the
walls of the sample after compression results in an increase in the conductivity from 10 -8 S cm -1
up to 0.002 S cm -1 as shown in Figure 5-10. This allows for another control parameter to achieve
167

a wide range of conductivity in these conductive porous devices. It also opens up the potential to
develop these devices as load bearing sensors.
5.5 Conclusions
This paper reports on the preparation of a novel 3D porous polymeric conducting device (PPCD)
derived from multi-percolated polymer blend systems. The work has focused on the preparation
of ultra low surface area porous substrates followed by the deposition of polyaniline conductive
polymer (PANI) on the internal porous surface using a layer-by-layer technique. In this way, the
percolation threshold of PANI in this porous conductive device can be reduced to a value as low
as 0.19%. Furthermore, depending on the amount of PANI deposited, the electrical conductivity
of the porous substrate can be controlled over several orders of magnitude from 10 -15 S cm -1 to
10 -3 S cm -1 .
Ternary and quaternary multi-percolated(hierarchically ordered) systems comprised of highdensity
polyethylene (HDPE), polystyrene (PS), poly(methyl methacrylate) (PMMA) and
poly(vinylidene fluoride) (PVDF) are melt-mixed and subsequently annealed in order to obtain
large interconnected phases. Subsequent extraction of PS, PMMA and PVDF within that blend
allows for the preparation of a fully interconnected porous HDPE substrate of ultra-low surface
area and highly uniform sized channels. This provides an ideal substrate for subsequent
polyaniline (PANI) addition. Using a layer-by-layer (LbL) approach, alternating poly(styrene
sulfonate) (PSS)/PANI layers are deposited on the internal surface of the 3-dimensional porous
polymer substrate. The PANI and sodium poly(styrene sulfonate) (NaPSS) both adopt an interdiffused
network conformation on the surface. The sequential mass deposition of PSS and PANI
multilayers has been studied in detail and an oscillating behavior is observed, due primarily to
the diffusion of PSS chains both in and out of the multilayer structure. Salt in the deposition
solution highly affects the polyelectrolyte construction by allowing for a more uniform
deposition and more thickly deposited PSS/PANI layers. The mass deposition growth for the
PSS/PANI system in all cases is not linear. Conductivity measurements show that the
conductivity of these samples increases from 10 -15 S cm -1 to 10 -5 S cm -1 as the number of
168

deposited PANI layers increases until a saturation plateau is reached at approximately 32 layers.
The conductive porous polymer device developed in this study is sensitive to applied
compression providing that the void volume percentage is sufficiently high. Higher loads result
in higher conductivity values with values as high as 10 -3 S cm -1 obtained. Although this approach
has been demonstrated here for an ultra-low surface area porous substrate, high surface area
substrates can also be readily prepared.
5.6 Acknowledgements
The authors express appreciation to the Natural Sciences and Engineering Research Council of
Canada for supporting this work. They also thank F. Normandin from the Institut des matériaux
industriels for his assistance with the conductivity measurements of the samples.
5.7 References
(1) Heywang, G.; Jonas, F. Advanced Materials 1992, 4, 116.
(2) Roman, L. S.; Andersson, M. R.; Yohannes, T.; Inganás, O. Advanced Materials 1997, 9,
1164.
(3) Kraft, A.; Grimsdale, Andrew C.; Holmes, Andrew B. Angewandte Chemie 1998, 110,
416.
(4) Schmidt, V. M.; Tegtmeyer, D.; Heitbaum, J. Journal of Electroanalytical Chemistry
1995, 385, 149.
(5) Wessling, B. Advanced Materials 1994, 6, 226.
(6) Contractor, A. Q.; Sureshkumar, T. N.; Narayanan, R.; Sukeerthi, S.; Lal, R.; Srinivasa,
R. S. Electrochimica Acta 1994, 39, 1321.
(7) Kaneto, K.; Kaneko, M.; Min, Y.; MacDiarmid, A. G. Synthetic Metals 1995, 71, 2211.
(8) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R.
H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539.
(9) Margolis, J. M. Conductive Polymers and Plastics; Chapman and Hall: New York, 1989.
(10) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.; Friend, R. H.; Moratti,
S. C.; Holmes, A. B. Nature 1995, 376, 498.
(11) Tessler, N.; Denton, G. J.; Friend, R. H. Nature 1996, 382, 695.
(12) Garnier, F.; Hajlaoui, R.; Yassar, A.; Srivastava, P. Science 1994, 265, 1684.
(13) Zasadzinski, J. A.; Viswanathan, R.; Madsen, L.; Garnaes, J.; Schwartz, D. K. Science
1994, 263, 1726.
(14) Maoz, R.; Frydman, E.; Cohen, S. R.; Sagiv, J. Advanced Materials 2000, 12, 424.
169

(15) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chemical
Reviews 2005, 105, 1103.
(16) Hammond, P. T. Advanced Materials 2004, 16, 1271.
(17) Decher, G.; Hong, J. D.; Schmitt, J. Thin Solid Films 1992, 210-211, 831.
(18) Paul, D. R.; Barlow, J. W. 1980, 18, 109
(19) Bourry, D.; Favis, B. D. Journal of Polymer Science Part B: Polymer Physics 1998, 36,
1889.
(20) Gubbels, F.; Jerome, R.; Teyssie, P.; Vanlathem, E.; Deltour, R.; Calderone, A.; Parente,
V.; Bredas, J. L. Macromolecules 1994, 27, 1972.
(21) Wang, Y.; Rubner, M. F. Macromolecules 2002, 25, 3284.
(22) Sumita, M.; Sakata, K.; Hayakawa, Y.; Asai, S.; Miyasaka, K.; Tanemura, M. Colloid &
Polymer Science 1992, 270, 134.
(23) Levon, K.; Margolina, A.; Patashinsky, A. Z. Macromolecules 2002, 26, 4061.
(24) Zhang, J.; Ravati, S.; Virgilio, N.; Favis, B. D. Macromolecules 2007, 40, 8817.
(25) Torza, S.; Mason, S. G. J. Coll. Interface Sci. 1970, 33, 67.
(26) Hobbs, S. Y.; Dekkers, M. E. J.; Watkins, V. H. Polymer 1988, 29, 1598.
(27) Zhenhua Yuan, B. D. F. AIChE Journal 2005, 51, 271.
(28) Sarazin, P.; Favis, B. D. Biomacromolecules 2003, 4, 1669.
(29) Decher, G.; Hong, J. D. European Conference on Organized Organic Thin Films 1991,
321.
(30) Clark, S. L.; Montague, M. F.; Hammond, P. T. Macromolecules 1997, 30, 7237.
(31) Shiratori, S. S.; Rubner, M. F. Macromolecules 2000, 33, 4213.
(32) Sui, Z.; Salloum, D.; Schlenoff, J. B. Langmuir 2003, 19, 2491.
(33) Ferreira, M.; Rubner, M. F. Macromolecules 1995, 28, 7107.
(34) Schoeler, B.; Kumaraswamy, G.; Caruso, F. Macromolecules 2002, 35, 889.
(35) Decher, G. Science 1997, 277, 1232.
(36) Caruso, F.; Caruso, R. A.; Mohwald, H. Science 1998, 282, 1111.
(37) Roy, X.; Sarazin, P.; Favis, B. D. Advanced Materials 2006, 18, 1015.
(38) Cheung, J. H.; Stockton, W. B.; Rubner, M. F. Macromolecules 1997, 30, 2712.
(39) Fou, A. C.; Rubner, M. F. Macromolecules 2002, 28, 7115.
(40) Kim, H. S.; Sohn, B. H.; Lee, W.; Lee, J. K.; Choi, S. J.; Kwon, S. J. Thin Solid Films
2002, 419, 173.
(41) Liang, Z.; Cabarcos, O. M.; Allara, D. L.; Wang, Q. Advanced Materials 2004, 16, 823.
(42) Ram, M. K.; Salerno, M.; Adami, M.; Faraci, P.; Nicolini, C. Langmuir 1999, 15, 1252.
(43) Ferreira, M.; Cheung, J. H.; Rubner, M. F. Thin Solid Films 1994, 244, 806.
(44) Dautzenberg, H.; Jaeger, W.; Kotz, J.; Phillip, B.; C., S.; Stscherbina, D. Carl Hanser
Verlag: Munich, Germany 1994.
(45) Harris, J. J.; Bruening, M. L. Langmuir 2000, 16, 2006.
(46) Dubas, S. T.; Schlenoff, J. B. Macromolecules 2001, 34, 3736.
(47) Braga, G. S.; Paterno, L. G.; Lima, J. P. H.; Fonseca, F. J.; de Andrade, A. M. Materials
Science and Engineering C 2008, 28, 555.
(48) Paloheimo, J.; Laakso, K.; Isotalo, H.; Stubb, H. Synthetic Metals 1995, 68, 249.
(49) Reignier, J.; Favis, B. D. Macromolecules 2000, 33, 6998.
(50) Yasuda, K.; Armstrong, R.; Cohen, R. Rheologica Acta 1981, 20, 163.
(51) Virgilio, N.; Favis, B. D.; Pepin, M.-F.; Desjardins, P.; L'Esperance, G. Macromolecules
2005, 38, 2368.
170

(52) Guo, H. F.; Packirisamy, S.; Gvozdic, N. V.; Meier, D. J. Polymer 1997, 38, 785.
(53) Ravati, S.; Favis, B. D. Macromolecules 2010, Submitted.
(54) Li, J.; Ma, P. L.; Favis, B. D. Macromolecules 2002, 35, 2005.
(55) Li, J.; Favis, B. D. Polymer 2001, 42, 5047.
(56) Klitzing, R. v.; Espert, A.; Asnacios, A.; Hellweg, T.; Colin, A.; Langevin, D. Colloids
and Surfaces A: Physicochemical and Engineering Aspects 1999, 149, 131.
(57) Picart, C.; Lavalle, P.; Hubert, P.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J. C.
Langmuir 2001, 17, 7414.
(58) Dubas, S. T.; Schlenoff, J. B. Langmuir 2001, 17, 7725.
(59) Tjipto, E.; Quinn, J. F.; Caruso, F. Journal of Polymer Science, Part A (Polymer
Chemistry) 2007, 45, 4341.
(60) Lavalle, P.; Gergely, C.; Cuisinier, F. J. G.; Decher, G.; Schaaf, P.; Voegel, J. C.; Picart,
C. Macromolecules 2002, 35, 4458.
(61) Schlenoff, J. B.; Dubas, S. T. Macromolecules 2001, 34, 592.
(62) Ha, B. Y.; Thirumalai, D. Macromolecules 1995, 28, 577.
(63) Klitzing, R.; Moehwald, H. Langmuir 1995, 11, 3554.
(64) Jaber, J. A.; Schlenoff, J. B. Langmuir 2007, 23, 896.
(65) Solonin, S. M.; Chernyshev, L. I. Soviet Powder Metallurgy and Metal Ceramics
(English translation of Poroshkovaya Metallurgiya) 1975, 14, 806.
(66) Montes, J. M.; Cuevas, F. G.; Rodriguez, J. A.; Herrera, E. J. Powder Metallurgy 2005,
48, 343.
(67) Lifshitz, E. M.; Landau, L. D.; Pitaevskii, L. P. Electrodynamics of Continuous Media,
1984; Vol. 8.
(68) Skorokhov, V. V. The Rheological Principles of Sintering Theory; Nauka: Kiev, 1972.
(69) Ravati, S., École Polytechnique de Montréal, 2010.
171

CHAPTER 6 - MORPHOLOGICAL STATES FOR A TERNARY POLYMER
6.1 Abstract
BLEND DEMONSTRATING COMPLETE WETTING
For the most part, ternary polymer blends demonstrate complete wetting behavior. Conceptually,
this is the state where one of the components will always tend to completely separate the other
two and from a thermodynamic viewpoint is described as the case where two of the three
possible binary spreading coefficients are negative and the other is positive, as defined by
Harkins spreading theory. This work examines the complete range of morphological states
possible for such a system over the entire ternary composition diagram as prepared by melt
mixing. A ternary polymer blend comprised of high-density polyethylene(HDPE),
polystyrene(PS), and poly(methyl methacrylate)(PMMA) is selected as a model system
demonstrating complete wetting and four sub-categories of morphologies can be identified
including: a) matrix/core-shell dispersed phase; b) tri-continuous; c) matrix/two separate
dispersed phases;, and d) bi-continuous/dispersed phase morphologies. Electron microscopy as
well as a technique based on the combination of focused ion beam irradiation and atomic force
microscopy are used to clearly illustrate and identify the various phases. Solvent
extraction/gravimetry is used to examine the extent of continuity of the systems so as to
effectively identify regions of high continuity. Triangular compositional diagrams are used to
distinguish these various morphological regions and the results are interpreted in light of the
interfacial tension of the various binary combinations and their subsequent spreading
coefficients. The effect of the molecular weight and of viscosity ratio on the phase size of the
various structures is also considered.
Keywords: ternary polymer blend, double-percolated morphology, tri-continuous, compositedroplet,
co-continuous structure
172

6.2 Introduction
It is well-known that the morphology of a polymer blend has a significant influence on its
properties(Paul & Newman, 1978; Utracki, 1998). For binary immiscible polymer blends, two
broad categories of morphology exist: the matrix/dispersed phase structure and the co-continuous
morphology (Figure 6-1a)(Favis, 2000; Potschke & Paul, 2003).
a)
b) c)
Figure 6-1. (a) schematic representation of the evolution of morphology in a binary blend, (b)
matrix/dispersed morphology, and (c) co-continuous morphology
173

The dispersed phase in a binary polymer blend can take the form of fibers, lamella and A-B-A
droplet-in-droplet type structures(Molau & Keskkula, 1966; Reignier & Favis, 2000). In an A/B
binary blend, the dispersed phase type structure(Figure 6-1b) is converted to a continuous type
structure through an increase in the composition. By increasing the concentration of dispersed
phase A, droplets coalesce resulting in a percolation threshold point being reached. This
percolation threshold is the first connected pathway in the blend(Figure 6-1a). Classic
percolation theory defines the percolation threshold as the onset of long-range connectivity in
random systems and it occurs at a volume fraction of 0.156 for a random mono-disperse
distribution of spheres(Scher & Zallen, 1970). Through a further increase of minor phase
concentration, levels of continuity increase until a fully-interconnected co-continuous
morphology is obtained(Figure 6-1c). Co-continuity is defined as the case where each phase is
fully continuous in the blend system. Since this often occurs over a concentration range for
binary polymer blends, this is also known as the region of dual-phase continuity. Phase inversion
is defined as the concentration point where co-continuity converts into a matrix/dispersed phase
morphology. It has been reported that the interfacial tension and the viscosity ratio can also
influence the position of the region of dual-phase continuity(Jordhamo, et al., 1986; Mekhilef &
Verhoogt, 1996).
Much less work has been carried out on the fundamental morphological states present in ternary
polymer blends comprised of significant quantities of 3 distinct phases. Recently, some papers
have described the morphological behavior of ternary systems with complex morphologies such
as A-B-C composite-droplet structures(Reignier & Favis, 2000; Reignier, et al., 2003) and
double-percolated morphology(Zhang, et al., 2007).
Complete wetting and partial wetting are two broad categories of morphological states
possible for ternary polymer blends. In an A/B/C system, complete wetting describes the case
where the most stable thermodynamic state is when one of the phases, say phase B, will always
position itself to completely separate phases A and C. In that case phases A and B completely
wet each other and phases B and C also completely wet each other. In the case of partial wetting,
the most stable thermodynamic state is when there is three-phase contact. In that case, for
example, droplets of B will situate at the A/C interface such that all three phases are in contact
with each other(Torza & Mason, 1970; Virgilio, Marc-Aurele, et al., 2009). Both complete and
partial wetting can be described by spreading theory as defined by Harkins equation. A
174

modified version of Harkins theory(Harkins & Feldman, 1922) was first reported by Torza et
al.(Torza & Mason, 1970). Hobbs et al.(Hobbs, et al., 1988) followed and developed a modified
Harkins equation and calculated three spreading coefficients of ternary polymer systems to
predict the possible morphological structures as shown in Equation 6-1 :
λ = γ − γ − γ
Equation 6-1 ij jk ik ij
where λij is defined as the spreading coefficient defining the tendency of component (i) to
encapsulate or spread onto component (j) in the matrix of component (k). The interfacial tensions
of the various polymer pairs are represented as γij, γik and γjk. In a certain ternary blend, 3
different spreading coefficients exist including λij = λik, λji = λjk, and λki = λkj.A positive value of
one of the spreading coefficients indicates that the ternary blend will demonstrate complete
wetting behavior as shown in Figure 6-2a. For example, a positive value of λik demonstrates that
phase (i) spreads over phase (k) leading to a separation of phases (j) and (k) by phase (i) at the
interface. A matrix/core-shell dispersed phase structure is an example of complete wetting in
which phase B, for example, encapsulates dispersed phase C in a matrix of A. On the other hand,
negative values for all three spreading coefficients indicates a partial wetting behavior in which
all components have an interface with each other, or all three meet along a common line of three
phase contact (Figure 6-2a)(Virgilio, Marc-Aurele, et al., 2009). A number of complex
morphologies in ternary HDPE/PS/PMMA blends have been investigated. Reignier et
al.(Reignier & Favis, 2000) examined the core-shell morphology of PMMA encapsulated by PS
in an HDPE matrix in detail. Zhang et al.(Zhang, et al., 2007) reported on the generation of a
double-percolated structure comprised of a PS layer at the HDPE/PMMA interface through the
control of the composition of the components. Virgilio et al.(Virgilio, Marc-Aurele, et al., 2009)
reported the case of partial wetting for a HDPE/PS/PP blend demonstrating a close-packed
droplet array of PS at the interface of HDPE and PP. Guo et al.(Guo, Gvozdic, et al., 1997; Guo,
Packirisamy, et al., 1997), studied the morphology of ternary and quaternary blends and
proposed a model which includes not only the interfacial tension, but also the interfacial area of
the phases. To date, however, the detailed examination of all the possible morphological states
over the entire composition range for a ternary polymer blend has not been carried out.
175

a)
λB/A > 0
b)
λB/A < 0
Figure 6-2. Schematic representation of (a) complete wetting: the case that one phase locates
between two other phases, and (b) partial wetting: the case that all phases are in contact with
each other
The above models employ thermodynamic properties in order to classify the morphology of a
ternary blend as complete or parial wetting. Rheological properties also play an important role,
particularly with respect to phase size and shape, most notably through viscosity ratio (Gupta,
1993; Luzinov, et al., 2000; Nemirovski, et al., 1995; Tchomakov, et al., 2004), and elasticity
ratio effects(Legros, et al., 1997; Vanoene, 1972).
The objective of this work is examined all the possible morphological states for a ternary
polymer blend, demonstrating complete wetting, over the entire composition range. A ternary
HDPE/PS/PMMA blend will be used as the model system and the morphology will be
determined through a combination of electron microscopy, FIB/AFM analysis as well as through
the estimation of continuity effects via selective solvent extraction/gravimetry of specific phases.
The observed morphologies will be related to spreading theory through a detailed examination of
the various interfacial tension values and spreading coefficients.
176

6.3 Experimental
6.3.1 Materials
Commercial homopolymers of HDPE, PMMA, and PS were used in this study to prepare ternary
samples. Low molecular weight poly(methyl methacrylate) and , high molecular weight
poly(methyl methacrylate) obtained from Sigma-Aldrich and the high-density polyethylene,
HDPE 3000 from Petromont. The L-PMMA was the base resin used in all parts of this work
while the H-PMMA was only used in the part discussing the effects of viscosity. Polystyrene, PS
615APR was purchased from Dow Chemical. Some of the characteristics of the homopolymers
are reported in Table 6-1.
The interfacial tension values and the Harkins spreading coefficients are tabulated in Table 6-2.
Table 6-1. Material Characteristics
Material
High Density
Polyethylene
Commercial
code
Mw * 10 -3
(g/mol)
Melt Index
(g/10min)
η0 *10 -3
at 190°C
(Pa.s)
η *×10 -3 at 25
rad s -1 (Pa.s)
Density
(g/cm 3 )
at 20 °C
177
Density
(g/cm 3 ) at
200 °C
3000 - 8.1 1.6 0.72 0.98 0.85
Polystyrene 615APR 290 - 5.9 1.5 1.04 0.97
L-Poly(methyl
methacrylate)
H-Poly(methyl
methacrylate)
- 12 - -- 0.04 1.19 1
182265 996 - --
7.8 1.2 1.1

Table 6-2. Interfacial Tension of Polymer Pairs from (Reignier & Favis, 2000) and Positive
Spreading Coefficient of Ternary Blends
6.3.2 Rheological Analysis
Interfacial tension
γ (mN/m)
Positive Spreading Coeff.
λ (mN/m)
γHDPE/PMMA = 8.6 λPS/PMMA = 2.6
γPS/PMMA = 2.4 λPMMA/HDPE = -5.9
γPS/HDPE = 5.1 λHDPE/PS = -11.3
The disc-shape samples were compression molded in the hot press at 200°C. A Bohlin constant
stress rheometer (CSM) in the dynamic mode equipped with 25 mm parallel disk geometry was
utilized to measure the linear viscoelastic properties of homopolymers containing 0.2%
antioxidant. All polymers were found to be stable at set temperature in the time sweep test for 8
minutes. After the determination of the viscoelastic region by stress sweep, a frequency sweep
allowed for the measure of the zero shear viscosity of each sample. The Carreau-Yasuda model
and IRIS Rheo-Hub 2008 software were employed to determine the zero shear viscosity of
PEMA. The complex viscosity of pure homopolymers is plotted as a function of frequency in
Figure 6-3. The average shear rate in the mixer is approximately 25 s -1 (Reignier, et al., 2003).
The zero shear viscosity and the viscosity at 25 s -1 for the various homopolymers are reported in
Table 6-1.
178

Complex voscosity (Pa.s)
1,0E+04
1,0E+03
1,0E+02
1,0E+01
0,01 0,1 1 10 100 1000
Figure 6-3. Complex viscosity as a function of frequency at 190ºC for homopolymers
6.3.3 Sample Preparation
Ternary polymer blends were melt-blended under a flow of dry nitrogen in a 30mL Plasti-Corder
internal mixer (Brabender) operating at 200 °C and 50 rpm for 8 minutes. To reduce thermal
oxidation of the polymers, 0. 2 wt % of Irganox B225 antioxidant from Ciba-Geigy was added to
the mixture. The mixing chamber was filled to 70% of its total volume. After mixing, the
samples were immediately cut from the mass and quenched in a cold water bath to freeze-in the
morphology. All concentrations have been reported based on component volume fraction.
6.3.4 Solvent Extraction
Frequency (rad s -1 )
In order to determine the continuity level of components in the blend, a solvent
extraction/gravimetric method was performed. It is a fast and straightforward technique to detect
HDPE
PS
179

the existence of continuous microstructures with a high accuracy when the components are
soluble in specific solvents. In this approach, the volumes of the components before and after
extraction are measured by weighing the sample and converting the weight to volume (Equation
6-2).
Equation 6-2
% Cocontinuity
=
∑Vinitial − ∑
n n
Vinitial and Vfinal are the volume of one component or present in the sample, before and after
extraction, calculated by weighing the sample and converting it to the volume. The reported
value is the average of %cocontinuity for several samples and the average error bar for high
continuity levels is ± 5% and for low continuity levels is ± 3%. Cyclohexane was utilized as
selective solvents for polystyrene and acetic acid for poly(methyl methacrylate).
6.3.5 Characterization of Phase Morphology
6.3.5.1 Microtoming and Scanning Electron Microscopy (SEM)
The specimens were cut and microtomed under liquid nitrogen to a plane face using a
microtome (Leica-Jung RM 2065) equipped with a glass knife and a cryochamber type (LN 21).
After the appropriate treatment of the sample with selective solvents (acetic acid for PMMA and
cyclohexane for PS) to remove one or both of the components, the sample surface was coated
with a gold/palladium by plasma deposition. A JEOL JSM 840 scanning electron microscope,
operated at a voltage of 10keV to 12keV, was used to obtain photomicrographs of the sample
surface.
∑
n
V
initial
V
final
180

6.3.5.2 Focused Ion Beam (FIB) and Atomic Force Microscopy (AFM)
After cryomicrotoming the specimens and coating them with a gold-palladium alloy, a Focused
Ion Beam (FIB), Atomic Force Microscopy (AFM) technique was used to examine the
HDPE/PS/PMMA blend. The surface of the samples was treated and etched using a Hitachi
focused ion beam FIB-2000A operating a 30 keV gallium beam. FIB uses a focused beam of
gallium ions which are accelerated to energies of 5-50 kiloelectronvolts. Using an electrostatic
lens, the ion beam can be focused on a very small spot, resulting in a high resolution because of
the small emitting area. In this work, a 0.8 nA beam current and a dwell time of 3 μs was applied
in order to remove approximately 3-4 μm of the surface. Milling was carried out parallel to the
observed surface. Since each polymer component has a different interaction with the gallium
beam, this approach induces topological differences and hence increases the contrast between
components(Virgilio, et al., 2005a). The milled surface of the specimen was then examined by a
scanning probe microscope dimension (AFM) with a Nanoscope IIIa controller in topological
mode. The atomic force microscope measures topography with a force probe. Silicon tips with
spring constant of 40 N/m and a resonant frequency of around 300 kHz was employed.
6.4 Results and Discussion
In ternary HDPE/PS/PMMA blends, the very high interfacial tension of PMMA and
HDPE(γHDPE/PMMA = 8.6 mN/m) and the low interfacial tension of PS/PMMA(γPS/PMMA = 2.4
mN/m) and HDPE/PS(γHDPE/PS = 5.1 mN/m) results in a positive spreading coefficient of PS over
PMMA (λPS/PMMA) with a value of 1.1 mN/m. The other two spreading coefficients are negative
and are λPMMA/PS = -5.9 mN/m and λHDPE/PMMA = -11.3 mN/m. This predicts a complete wetting
case with the development of a thermodynamically stable PS layer between the PE and PMMA
as represented for a ternary 40/20/40 HDPE/PS/PMMA blend as shown in Figure 6-4. In Figure
6-4, the various phases can be clearly identified by the topographical heights induced by FIB
etching and subsequently quantified by AFM analysis in the topographical mode. Previous
works(Reignier & Favis, 2000, 2003a, 2003b; Reignier, et al., 2003) have shown several submorphologies
of HDPE/PS/PMMA such as matrix/core-shell morphology and double-percolated
181

morphology(Zhang, et al., 2007). In the next sections it will be shown that four
thermodynamically stable sub-classes of morphologies exist for HDPE/PS/PMMA depending on
the composition of phases.
Figure 6-4. FIB-AFM image of complete wetting morphology for ternary 40/20/40
HDPE/PS/PMMA blend showing the layer of PS at the interface of HDPE and PMMA. Scan size
is 25μm×25μm. The bar to the right of the FIB/AFM micrograph indicates the colours associated
with the topographical height in nm. The white line in the image indicates the section analyzed.
6.4.1 Matrix/Core-Shell Dispersed Phase Morphology
The matrix/core-shell dispersed phase morphology is a well-known sub-morphology for ternary
blends. This morphology in a ternary A/B/C blend constitutes droplets of phase A encapsulated
by phase B in the matrix of phase C (shown by schematic in Figure 6-5a). A micrograph of this
morphology is shown for a ternary 60/10/30 HDPE/PS/PMMA blend comprised of a matrix of
HDPE with the PMMA phase present as a core in a PS shell phase(Figure 6-5b).
50
0
-50
PMMA
Section
PS
1 2
Distance μm
HDPE
182

a)
c)
b)
d)
e) f)
PMMA
PMMA
HDPE
HDPE
183

g)
Figure 6-5. (a) Schematic representation of the matrix/core-shell dispersed phase morphology,
(b) SEM micrograph of 60/10/30 HDPE/PS/PMMA blend after extraction of PS by cyclohexane,
(c) schematic representation of tri-continuous morphology (type I), (d) FIB-AFM image of
40/20/40 HDPE/PS/PMMA showing tri-continuous morphology. Scan size is 25μm×25μm. The
bar to the right of the FIB/AFM micrograph indicates the colours associated with the
topographical height in nm, (e) schematic representation of tri-continuous(type II), (f) SEM
micrograph of 30/30/40 HDPE/PS/PMMA after extraction of PS by cyclohexane, (g) schematic
representation of matrix/two separate dispersed phases morphology, and (h) SEM micrograph of
10/80/10 HDPE/PS/PMMA microtomed.
Note that in all schematics black represents the HDPE phase, grey represents the PS phase and
white represents the PMMA phase.
6.4.2 Tri-continuous Morphology
h)
In a ternary blend with a matrix/core-shell morphology, increasing the concentration of
the core phase results in the coalescence of core phases which leads to the formation of a tricontinuous
morphology. In the model ternary HDPE/PS/PMMA blend studied here, as predicted
184

y Harkins theory, PS will always situate at the interface. Thus, the coalescence of the PMMA
cores directly results in the coalescence of PS shells to each other. When the concentration of the
PMMA phase becomes high enough to create a continuous network throughout the sample, the
PS phase is forced to adopt a continuous form to spread over the PMMA phase(shown by
schematic and by micrograph in Figures 6-5c and 6-5d). FIB-AFM characterization of 40/20/40
HDPE/PS/PMMA blend reveals approximately such a morphology(Figure 6-5d). This is a tri-
continuous structure or double-percolated morphology and by definition all phases are highly
continuous.
Tri-continuous structures can be classified into two categories: type I and type II. The former is
shown in Figure 6-5c) and 6-5d), as described above. In that case the PS composition is low
enough such that most of the PS situates at the interface, apart from some minor droplet
formation in the PMMA. In the Type II case, as the concentration of PS is increased, more
isolated droplets are located in the PMMA since the HDPE/PS/PMMA interface is already
saturated by PS. These droplets coalesce into elongated structures as the concentration increases
until the PS within the PMMA also forms a fully continuous interconnected phase. Such a
structure is shown by micrograph and by schematic in Figures 6-5e) and 6-5f) Note that the
structure can be described as a co-continuous morphology of PS and PMMA separated by a PS
layer from a fully continuous HDPE phase. It should be noted that this morphology is still
consistent with Harkins theory as phase B still fully separates phases A and C. In Figure 6-5f) for
30/30/40 HDPE/PS/PMMA the voids left by the selective extraction of PS can be clearly seen to
show that PS forms an interlayer between HDPE and PMMA. A fully interpenetrated network of
PS and PMMA is also visible.
Hence, tri-continuous type I is transformed to tri-continuous type II by increasing the
concentration of the phase separating the other two phases.
6.4.3 Matrix/ Two Separate Dispersed Phases Morphology
Another type of sub-morphology in the complete wetting case of HDPE/PS/PMMA is the matrix
with two separate dispersed phases morphology(shown as a schematic in Figure 6-5g). In that
185

case the grey matrix is PS, with dispersed droplets of PMMA (white) and separate dispersed
droplets of HDPE (black). Figure 6-5h shows a micrograph of droplets of PMMA and HDPE
phases in a PS matrix. Since it is hard to distinguish PMMA and HDPE droplets, the PMMA is
selectively extracted showing voids where the PMMA was located and clearly indicating the
presence of HDPE droplets.
6.4.4 Bi-Continuous/Dispersed Phase Morphology
The last case of sub-morphology for HDPE/PS/PMMA is the bi-continuous/dispersed phase
morphology comprised of a co-continuous structure for two phases with the third phase present
as a dispersed phase. This case is generated when the concentration of the middle separating
phase increases resulting in a co-continuous structure of the middle phase with either of the other
two phases. The droplets of the third phase are distributed in the middle separating phase (shown
as a schematic in Figure 6-6a). Two examples of such a morphology are shown in Figures 6-6b)
and 6-6c). 50/40/10 HDPE/PS/PMMA is shown in 6-6b) and it can be seen that the HDPE and
PS phases are continuous with droplets of PMMA in the PS phase. 50/25/25 HDPE/PS/PMMA is
shown in Figure 6-6c).
In a ternary blend with a bi-continuous/dispersed phase structure increasing the amount of inner
dispersed phase and decreasing the extent of continuous outer phase results in an inversion in
morphology. In Figures 6-6 d), e) and f), PMMA and PS have a co-continuous structure and
HDPE is present as droplets. This inversion can be shown in a comparison of 50/25/25
HDPE/PS/PMMA(Figure 6-6c) where droplets of PMMA can be seen in the PS phase which, in
turn, is co-continuous with HDPE, and 10/30/60 HDPE/PS/PMMA(Figure 6-6e) where droplets
of HDPE can be seen in PMMA which is co-continuous with PS. The PS phase is extracted in
Figure 6-6f for the 10/30/60 HDPE/PS/PMMA clearly showing numerous droplets of HDPE
distributed in the PS phase.
Figure 6-7 represents the sub-morphologies in the complete wetting case that can be converted to
each other by varying the phase composition.
186

a) b) c)
d) e) f)
PS
Figure 6-6. a) Schematic representations Distance
of bi-continuous/dispersed phase morphology with
small droplets of white phase, b) FIB-AFM image of 50/40/10 HDPE/PS/PMMA showing bicontinuous/dispersed
phase morphology with small droplets of PMMA. Scan size is 8μm×8μm,
(c) FIB-AFM image of 50/25/25 HDPE/PS/PMMA showing bi-continuous/dispersed phase
morphology with droplets of PMMA. Scan size is 4μm×4μm, d) schematic representations of bicontinuous-dispersed
phase morphology with small droplets of black phase, e) FIB-AFM image
of 10/30/60 HDPE/PS/PMMA showing bi-continuous/dispersed phase morphology with
droplets of HDPE. Scan size is 11.7μm×11.7μm, and f) SEM micrograph of 10/30/60
HDPE/PS/PMMA after extraction of PS. Small droplets of HDPE are distributed in PS
Note that in the schematics, black represents the HDPE phase, grey represents the PS phase and
white represents the PMMA phase.
6.4.5 Triangular Composition/Morphology Diagram
Thirty different samples comprised of various compositions of HDPE/PS/PMMA (Figure 6-8a))
were prepared to better understand the composition dependance of the various morphological
187

states for complete wetting and their inter-conversion. The various morphological states
observed are presented in Figures 6-8b) and 6-8c).
A triangular composition/morphology diagram (Figure 6-8c) was prepared. The sides of the
triangle show the volume fraction of HDPE, PS, and PMMA phases ranging from 0% to 100%.
The morphologies shown in Figures 6-8b) and 6-8c) will typically be observed in all ternary
polymer blends demonstrating complete wetting. However, it should be noted that the boundaries
between regions, as shown in Figure 6-8c), could also be influenced by the viscosity of the
various phases and this point will be examined in future work. Figure 6-8c) demonstrates the
concentration regimes for: a) tricontinuous blends of both type I and type II behavior (Cases I
and II in Figure 6-8); b) matrix/separate dispersed phases with PS as matrix phase (Case III); bi-
continuous/dispersed phase with HDPE as droplets (Case IV) or with PMMA as droplets (Case
V); and matrix/core-shell either with PMMA as matrix (Case VI) or with HDPE as matrix (Case
VII). These regions were detected by a combination of both SEM and FIB-AFM and through the
use of selective extraction/gravimetric analysis as quantitative technique to detect inversion
points.
Phase (A): black
Phase (B): grey
Phase (C): white
Figure 6-7: Evolution of the morphology in a ternary A/B/C blend demonstrating complete
wetting as a function of the concentration of phases B and C. Phase A concentration is held
constant.
188

a)
b)
HDPE : black PS : grey PMMA : white
189

c)
Figure 6-8. a) Triangular concentration diagram showing the composition of the various ternary
HDPE/PS/PMMA blends examined in this study, b) various morphological states for ternary
HDPE/PS/PMMA, and c) triangular concentration diagram showing the various regions of in the
morphological states for ternary HDPE/PS/PMMA
190

6.4.6 Continuity, Co-continuity, and Tri-continuity of Phases
One very important issue in the selection of a ternary blend for a particular application is the
continuity level of each phase. In this part of the work the effect of the composition of phases on
continuity levels is examined. The same 30 samples examined in the previous section above for
morphology are also examined here for their quantitative PS and PMMA continuity. It was not
possible to obtain the continuity of the HDPE phase since HDPE could not be selectively
extracted. Triangular continuity plots are introduced(Figures 6-9a and 6-9b) to show the
continuity of the PS and PMMA phases respectively. In this HDPE/PS/PMMA system, the
continuity levels of PS and PMMA phases are identified through the selective solvent
extraction/gravimetry of PMMA by acetic acid(Figure 6-9a) and of PS by cyclohexane(Figure 6-
9b). The axes in Figures 6-9a and 6-9b represent the volume fraction of PMMA and PS. Note
that the volume fraction of HDPE for these plots can be estimated by subtracting the combined
volume fractions of PS and PMMA from 100%. For example, in the sample comprising 40%PS
and 40%PMMA the remaining 20% would be HDPE. Figures 6-9a and 6-9b reveal the various
continuity levels for both PS and PMMA in the ternary system. These diagrams show that
increasing the concentration of phases results in increased continuity levels until that phase
becomes the matrix and continuity achieves 100%. In Figure 6-9b there are two regions with a
100% continuity level of the PS phase representing tri-continuous morphologies type I (low PS
concentration) and type II (high PS concentration).
In a binary blend the continuity percolation threshold is only one point. In a ternary blend it
becomes a line due to the simultaneously changing concentration of two of the phases in the
blend. The percolation threshold lines for PS and PMMA are represented in both Figures 6-9a
and 6-9b where the colour changes from purple to blue.
The data from the continuity diagrams for PS and PMMA in Figures 6-9a) and 6-9b), can be
combined with the morphological study shown in the triangular diagram of Figure 6-8c) in
order to produce even more well defined continuity boundaries for the various morphological
states. The new 3-axes triangular diagram is shown in Figure 6-9c). Since it is impossible to
selectively extract HDPE without also extracting PS and PMMA, the upper border of region
191

IV cannot be accurately determined by quantitative continuity measurements and remains
defined by morphological observation. All the various morphological states are presented.
a)
b)
Percolation
threshold lines
Percolation
threshold line
192

c)
V
III
II
I
VII
I
IV
VI
Figure 6-9. Continuity diagrams for ternary HDPE/PS/PMMA blends. a) quantitative continuity
level of PMMA (percolation threshold line is shown in black), b) quantitative continuity level of
PS (percolation threshold line is shown in black), and c) triangular concentration diagram
showing the various regions of in the morphological states for ternary HDPE/PS/PMMA after
correction with the PS and PMMA continuity data.
193

Region I is divided into two regions since in some parts of region I in Figure 6-8c, the
continuity of the PS phase is lower than 80%(around 60%). This additional region of lower
continuity is shown as region VII in Figure 6-9 and is termed the tricontinuous type I-type II
transition region. It represents the case where there is a PS layer at the interface of HDPE
and PMMA with numerous PS droplets within the PMMA.
6.4.7 Morphology Evolution due to Concentration Ratio of Two Components
in Complete Wetting Case
Figure 6-10 indicates two compositional lines, or schemes, across the 3 axes triangular
diagram which traverse multiple morphological states as shown in Figure 6-9c).
Scheme (a) in Figure 6-10 is the case where the PMMA concentration is held at 40% while
the combined concentration of HDPE and PS is always held at 60%. This scheme allows one to
study the morphology development of the PS phase in the ternary HDPE/PS/PMMA blend at
different composition ratios of HDPE to PS. In scheme a, the effect of the concentration of
the inner(HDPE) and middle(PS) phases on the morphology of the middle phase is
examined. The morphology development is quantified and supported by the PS continuity
data obtained via selective extraction/gravimetric measurements.
194

Continuity Scheme (a)
Figure 6-10. Triangular concentration diagram showing two pathways: Continuity Scheme (a),
and Continuity Scheme (b). Scheme (a) is shown in Figure 6-11 and represents the case where
the concentration of PMMA is kept constant at 40% of total while the concentration ratio of the
PS and HDPE phases change. Scheme (b) is shown in Figure 6-12 where the concentration of
PS is kept constant at 10% of total while the concentration ratio of PMMA and HDPE phases
change. These two pathways were chosen since they pass through a range of different
morphological states.
In Figure 6-11, the continuity of the PS phase following Scheme a is examined as a function of
the % concentration of HDPE in (HDPE plus PS) where HDPE plus PS always equals 60% in
the ternary blend with PMMA. Thus 100% HDPE refers to the 60%HDPE/40%PMMA blend
while 50% HDPE in Figure 6-11 refers to the 30%HDPE/30%PS/40%PMMA blend. At 0% of
HDPE a co-continuous morphology of 40/60 PMMA/PS is formed (point a in Figure 6-11).
195
Continuity Scheme (b)

Continuity (%) of PS phase
a)
100
90
80
70
60
50
40
30
20
10
0
a
b
0 10 20 30 40 50 60 70 80 90 100
Volume fraction of HDPE(HDPE%+PS% = 60% of total)
c
b) c)
d) e) f)
Figure 6-11. PS continuity and the corresponding morphology as a function of HDPE
concentration following Scheme (a) from Figure 6-10. The PMMA phase concentration is held
constant at 40% volume fraction in all samples. In the schematics, green is the PMMA phase, red
is the PS phase and blue is the HDPE phase.
d
e
f
196

Addition of a low concentration of HDPE to the system leads to the encapsulation of HDPE
droplets in a highly continuous PS network resulting in a bi-continuous/dispersed phase
morphology and a slight reduction of PS continuity(point b in Figure 6-11). A further increase of
HDPE phase generates a tri-continuous morphology(type II) where the coalescence of HDPE
droplets forms a continuous phase of HDPE in highly continuous PS (point c in Figure 6-11).
Note that at point c for the 30/30/40 HDPE/PS/PMMA blend, the continuity level of PS is
lowered to 68% Such a decrease in the continuity level is attributed to the entrapment of some
parts of the PS phase in the PMMA network. Point d indicates the expected drop in continuity
after the formation of the type I tricontinuous structure at point e. At point e, the 82% HDPE
volume fraction corresponds to the 50/10/40 HDPE/PS/PMMA blend. This type I tri-continuous
structure with a PS continuity of 76% represents the case where 50% HDPE and 40% PMMA
form a co-continuous structure with a thin PS layer situated at the HDPE/PMMA interface.
Finally, replacing all the PS with HDPE results in a (60%HDPE matrix/40%PMMA dispersed
phase) blend (point f in Figure 6-11).
Local maximum continuity of 78% is obtained for 48/12/40 HDPE/PS/PMMA blend from Figure
6-11. The most important observation for scheme a, where the PMMA concentration is held at
40%, is that due to the thermodynamic limitations of spreading, the system is never able to form
PS droplets even at a very low concentration of PS. PS always spreads between HDPE and
PMMA in order to separate those two components. Thus even the smallest amount of PS in
scheme a) presents a layer structure as opposed to droplets.
In scheme (b), from Figure 6-10, the concentration of the PS phase is kept constant at 10%
volume fraction and the concentration ratio of the two other phases changes. In Figure 6-12, the
continuity of the PS phase following Scheme b is examined as a function of the % concentration
of HDPE in (HDPE plus PMMA) where HDPE plus PMMA always equals 90% in the ternary
blend with PMMA. Thus, 100% HDPE refers to the 90%HDPE/10%PS blend, while 50% HDPE
in Figure 6-12 refers to the 45%HDPE/10%PS/45%PMMA blend. This scheme allows for the
study of the effect of the composition of the inner phase(HDPE) and the outer phase(PMMA)
on the morphology of the middle phase(PS). It is interesting to note that in this case, the PS
continuity curve is almost symmetric. At 0% HDPE, the continuity of PS is zero due to the
presence of PS droplets within the matrix of PMMA(point a in Figure 6-12).
197

Continuity (%) of PS phase
100
90
80
70
60
50
40
30
20
10
0
a
b
0 10 20 30 40 50 60 70 80 90 100
Volume fraction of HDPE(HDPE%+PMMA% = 90% of total)
a) b)
c) d) e)
Figure 6-12. PS continuity and the corresponding morphology as a function of HDPE
concentration following Scheme (b) from Figure 6-10. The PS phase concentration is held
constant at 10% volume fraction in all samples. In the schematics, green is the PMMA phase, red
is the PS phase and blue is the HDPE phase.
c
d e
198

Addition of small amounts of HDPE to the binary blend of PS/PMMA results in a matrix/core-
shell dispersed phase morphology where the PS is the shell and HDPE is the core(point b in
Figure 6-12). Further increasing the concentration of HDPE leads to an increase in both the
number of core-shell dispersions and the core volume. At 33% HDPE, the continuity data shows
that HDPE forms its first connected pathway, the percolation threshold for HDPE. A further
increase of HDPE to 44% results in the formation of a tri-continuous (type I) morphology as
shown at point c where all phases, HDPE, PS and PMMA, are continuous. Figure 6-12 indicates
that, at the peak of the continuity diagram, a perfectly uniform and continuous layer of 10%PS
with a continuity of 97% situates at the interface of co-continuous HDPE and PMMA phases. In
a previous paper(J. Zhang, et al., 2007) from this laboratory, a PS continuity level of 69% was
observed for a 3%PS/50%HDPE/47%PMMA blend. Figure 6-12 suggests that the maximum
continuity for PS is obtained at a blend concentration of 10PS/43HDPE/47PMMA. Further
increasing the quantity of HDPE at point d, results in droplets of PMMA encapsulated in PS
shells. Finally, at 100%HDPE a matrix/droplet morphology of 90/10 HDPE/PS is observed(point
e in Figure 6-12).
6.4.8 Observations on the Effect of Viscosity on the Morphological State
It is well known that the viscosity ratio can have a significant effect on both the size and shape of
dispersed phases in a polymer blend(B. D. Favis, 2000). In this section it will be shown that
extreme changes in the viscosity of the PMMA phase (L-PMMA and H-PMMA) in a
HDPE/PS/PMMA blend of constant composition still results in morphological structures where
the PS separates HDPE and PMMA. Note that L-PMMA is the same PMMA used in all the rest
of this work. The viscosity of high molecular weight PMMA used here is almost 200 times that
of the viscosity of low molecular weight PMMA leading to a huge corresponding increase in the
PS/PMMA viscosity ratio. Two ternary blends comprised of 35/40/25 HDPE/PS/L-PMMA and
35/40/25 HDPE/PS/H-PMMA are prepared and the resulting morphology is shown in Figure 6-
13. Both blends using low molecular weight PMMA and high molecular weight PMMA
demonstrate a bi-continuous/PMMA droplet morphology, Figures 6-13a and 6-13b show that the
L-PMMA phase is present as numerous small droplets encapsulated in PS. In Figure 6-13a the L-
199

PMMA is selectively extracted while in Figure 6-13b, only the PS is extracted. The micrographs
clearly show a bicontinuous morphology of HDPE and PS with L-PMMA dispersed as droplets
in the PS phase. PS separates the HDPE and L-PMMA. In this case, PS interpenetrates into the
L-PMMA phase rather than the HDPE phase due to lower interfacial tension of PS/PMMA and
the low viscosity of the L-PMMA.
a)
c) d) e)
Figure 6-13. SEM micrograph of morphology of (a) 35/40/25 HDPE/PS/L-PMMA after
extraction of L-PMMA by acetic acid, (b) 35/40/25 HDPE/PS/L-PMMA after extraction of PS
by cyclohexane, (c) 35/40/25 HDPE/PS/H-PMMA(cryofracture), (d) 35/40/25 HDPE/PS/H-
PMMA after extraction of PMMA by acetic acid, and (e) 35/40/25 HDPE/PS/H-PMMA after
extraction of PS by cyclohexane,
In Figures 6-13c) and d), H-PMMA is used instead of L-PMMA and the PS is selectively
extracted to indicate its position. In that case, H-PMMA is present as a few large droplets
encapsulated in thin PS shells(see Figure 6-13c). Figure 6-13d) shows that the PS forms a
b)
200

icontinuous network with the HDPE. Thus, in both the L-PMMA case and the H-PMMA case,
the morphological classification of bicontinuous/dispersed PMMA phase is unchanged.
Nevertheless, the scale of the resulting morphological structures has been significantly modified
by the viscosity of the PMMA as the system attempts to respond to the dramatically reduced
interfacial area of the H-PMMA droplets. In addition, it is more difficult for the PS to penetrate
the H-PMMA (as compared to L-PMMA) due to its significantly higher viscosity. Thus, the use
of H-PMMA results in a significant increase in the interfacial area between HDPE and PS
phases. A comparison of the thickness of the continuous HDPE branches in the L-PMMA blend
as compared to that in the H-PMMA blend clearly show that the HDPE branches are
significantly thicker in the L-PMMA case. This unambiguously supports the conclusion of an
increased HDPE/PS interfacial area for the H-PMMA blend.
The surface free energy for a blend system is given by the following equation:
Equation 6-3 ∑
where G is the Gibbs free energy, represents the interfacial tension between phases i and j,
and is the interfacial area between phases i and j. Since the system is unable to dissipate the
applied mechanical energy applied during mixing by generating a high interfacial area between
H-PMMA and PS, it must compensate by generating an increased interfacial area between the
HDPE and PS phases (Figure 6-13g).
6.4.9 Factors Influencing Phase Sizes
It is well known that the composition, interfacial tension, and viscosity ratio are three of the most
important factors influencing the phase size in binary polymer blends (A. K. Gupta, 1993; Guo,
Packirisamy, et al., 1997; Hobbs, et al., 1988; Nemirovski, et al., 1995) . It will be shown in this
section that these factors also play an important role in determining the phase size in ternary
polymer blends.
201

a) b)
c)
PS
HDPE
PMMA
Figure 6-14. FIB-AFM topographic surface image of HDPE/PS/PMMA at various compositions,
(a) 50/40/10, (b) 50/25/25, and (c) 10/30/60. The white lines in the images indicate the sections
analyzed for each blend.
202

Three examples are shown in Figure 6-14: 50/40/10 HDPE/PS/PMMA, 10/30/60
HDPE/PS/PMMA, and 50/25/25 HDPE/PS/PMMA. PMMA has a much lower viscosity than PS
which has a lower viscosity than HDPE. Also, the PS/PMMA interfacial tension is much lower
than the PS/HDPE or PMMA/HDPE one (see Table 6-2). In the case of 50/40/10
HDPE/PS/PMMA(Figure 6-14a), the PMMA phase is at a low concentration (10%), low
PMMA/PS viscosity ratio and a low PMMA/PS interfacial tension (see Tables 6-1 and 6-2). It
can be seen that the PMMA is distributed in the PS phase as numerous small PMMA droplets
leading to high interfacial area between PS/PMMA. In that same sample, PS and HDPE form a
co-continuous structure. The FIB/AFM micrographs allow for a clear identification of the small
PMMA droplet phases in the PS. The topographical heights following the trace line show the
short interval between PS and PMMA phases and the long interval for the HDPE phase. In the
ternary 50/25/25 HDPE/PS/PMMA blend, the increased concentration of PMMA results in the
coalescence of PMMA droplets resulting in the formation of large PMMA droplets. In the case
of the ternary blend comprised of 10/30/60 HDPE/PS/PMMA large droplets of HDPE are
observed in the PS phase while PS and PMMA comprise a co-continuous structure(Figure 6-
14c). These results are consistent with the low viscosity ratio and interfacial tension of
PS/PMMA and high viscosity ratio and interfacial tension of HDPE/PS.
6.5 Conclusions
Complete wetting in a ternary polymer blend is an interfacial tension driven state where one of
the components will always tend to completely separate the other two. It occurs when one of the
three possible binary spreading coefficients in the system has a positive value, as defined by
Harkins spreading theory. This work examines the complete range of morphological states
possible for such a system over the entire ternary composition diagram as prepared by melt
mixing. High-density polyethylene(HDPE), polystyrene(PS), and poly(methyl
methacrylate)(PMMA) are selected as a model system showing a positive spreading coefficient
of PS over PMMA, thus in all cases the PS phase separates HDPE and PMMA. Four subcategories
of morphologies can be identified, depending on the composition of phases, including:
a) matrix/core-shell dispersed phase; b) tri-continuous; c) bi-continuous/dispersed phase;, and d)
203

matrix/two separate dispersed phases morphologies. The phases in these sub-morphologies are
identified and illustrated qualitatively by electron microscopy as well as a technique based on the
combination of focused ion beam irradiation and atomic force microscopy. Both qualitative and
quantitative data are shown in triangular compositional diagrams to distinguish these various
morphological regions and the results are interpreted in light of the interfacial tension of the
various binary combinations and their subsequent spreading coefficients. Solvent
extraction/gravimetry is used as quantitative experiment to confirm the morphological
observations and to examine the extent of continuity of the systems so as to effectively identify
regions of high continuity. It is shown that the viscosity of the phases can have a significant
influence on the interfacial area generated between the phases in the blend system, but does not
effect the classification or type of morphological state which appears to be governed solely by
spreading theory.
6.6 Acknowledgement
The authors express appreciation to the Natural Sciences and Engineering Research Council of
Canada for supporting this work.
6.7 References
(1) Paul, D.; Newman, S. Polymer Blends; Academic Press: New York, 1978; Vol. 1, 2.
(2) Utracki, L. A. Commercial Polymer Blends Springer - Verlag, 1998.
(3) Favis, B. D. In Polymer Blends: Formulation and Performance, Two-Volume Set; Paul,
D. R., Bucknall, C. B., Eds.; John Wiley & Sons, Inc: New York, 2000; Vol. 1, p 239.
(4) Potschke, P.; Paul, D. R. Journal of Macromolecular Science - Polymer Reviews 2003,
43, 87.
(5) Molau, G. E.; Keskkula, H. Journal of Polymer Science Part A-1: Polymer Chemistry
1966, 4, 1595.
(6) Reignier, J.; Favis, B. D. Macromolecules 2000, 33, 6998.
(7) Scher, H.; Zallen, R. The Journal of Chemical Physics 1970, 53, 3759.
(8) Jordhamo, G. M.; Manson, J. A.; Sperling, L. H. Polymer Engineering and Science 1986,
26, 517.
204

(9) Mekhilef, N.; Verhoogt, H. Polymer 1996, 37, 4069.
(10) Reignier, J.; Favis, B. D.; Heuzey, M.-C. Polymer 2003, 44, 49.
(11) Zhang, J.; Ravati, S.; Virgilio, N.; Favis, B. D. Macromolecules 2007, 40, 8817.
(12) Torza, S.; Mason, S. G. Journal of Colloid and Interface Science 1970, 33, 67.
(13) Virgilio, N.; Marc-Aurele, C.; Favis, B. D. Macromolecules 2009, 42, 3405.
(14) Harkins, W. D.; Feldman, A. Journal of the American Chemical Society 1922, 44, 2665.
(15) Torza, S.; Mason, S. G. J. Coll. Interface Sci. 1970, 33, 67.
(16) Hobbs, S. Y.; Dekkers, M. E. J.; Watkins, V. H. Polymer 1988, 29, 1598.
(17) Guo, H. F.; Gvozdic, N. V.; Meier, D. J. Polymer 1997, 38, 4915.
(18) Guo, H. F.; Packirisamy, S.; Gvozdic, N. V.; Meier, D. J. Polymer 1997, 38, 785.
(19) A. K. Gupta, K. R. S. Journal of Applied Polymer Science 1993, 47, 167.
(20) Nemirovski, N.; Siegmann, A.; Narkis, M. Journal of Macromolecular Science - Physics
1995, B34, 459.
(21) Luzinov, I.; Pagnoulle, C.; Jerome, R. Polymer 2000, 41, 7099.
(22) Tchomakov, K. P.; Favis, B. D.; Huneault, M. A.; Champagne, M. F.; Tofan, F. Polymer
Engineering and Science 2004, 44, 749.
(23) Vanoene, H. Journal of Colloid and Interface Science 1972, 40, 448.
(24) Legros, A.; Carreau, P. J.; Favis, B. D.; Michel, A. Polymer 1997, 38, 5085.
(25) Virgilio, N.; Favis, B. D.; Pepin, M.-F.; Desjardins, P.; L'Esperance, G. Macromolecules
2005, 38, 2368.
(26) Reignier, J.; Favis, B. D. Polymer 2003, 44, 5061.
(27) Reignier, J.; Favis, B. D. AIChE Journal 2003, 49, 1014.
(28) Favis, B. D. In Polymer Blends: Formulation and Performance; Paul, D. R., Bucknall, C.
B., Eds.; John Wiley & Sons: John Wiley & Sons, 2000; Vol. I, p 239.
205

CHAPTER 7 - GENERAL DISCUSSIONS AND PERSPECTIVE
In the previous chapters, it was shown that a hierarchically ordered, interfacial tension driven
morphology can be assembled. In this multiple-percolated structure, thermodynamic properties
can determine the number of interfaces between phases(the number of contacts between phases).
In the next perspective of the work the influence of viscous forces on the morphology by
changing the shape and size of the phases can be studied. For example, by increasing the shear
stress in a bi-continuous/dispersed phase morphology, the size of the droplets should be
decreased due to the increasing breakup rate as compared to coalescence. Moreover, due to the
higher shear force, the shape of the droplets may also become more elongated. Therefore,
changing the applied shear stress on the system would be expected to result mostly in the
variation of the size and shape of the phases. Changing the hierarchical order of phases in the
blend, which depends significantly on the thermodynamic properties of the system, is almost
impossible. It is well-known that the third factor affecting the morphology of multi-component
blends is the composition of phases. This factor affects the morphology by imposing a
geometrical restriction between phases and controls the rate of breakup and coalescence. At high
concentration of one phase the droplets of that phase coalesce further with each other and
therefore the rate of coalescence increases until a continuous structure is constructed. For blends
comprised of more than three components, composition can also play a role in the prediction of
the phase morphology. Thermodynamics plays another role in the morphology of ternary blends
through the governing Harkins equation and the prediction of three-phase contact and two-phase
contact morphologies as mentioned in detail in chapter 6. In both cases of complete and partial
wetting, a high interfacial tension between phases results in a decrease of the interfacial area
between them leading to larger phase sizes. All the effects of these three
parameters(composition, interfacial tension, and shear stress) are schematically shown for a
complete wetting case of a ternary blend in Figures 7-1 and 7-2. Figures 7-1 and 7-2 illustrate
various morphologies of a blend with a matrix of phase (A) depending on the interfacial tension
of B/C and the composition of phases B and C at low shear rate and high shear rate, respectively.
It is shown that at low concentration of minor phases, a matrix/core-shell morphology is
observed. At higher concentration of minor phases, tri-continuous and bi-continuous/dispersed
206

structures are formed due to the coalescence of droplets. At low interfacial tension of B/C the
interfacial area between these phases increases and small droplets of phase (C) are dispersed in
phase (B). For the case of a high interfacial tension of B/C, the droplets coalesce with each other
and large droplets or a continuous structure of that phase is observed. Finally the effect of shear
rate is shown and at low shear rate, the shape of the phases are more regular and the phase sizes
are larger due to the higher capillary number. At high shear rate, the structures become more
elongated and the number of composite droplets and continuous islands increase due to the
smaller size of the phases.
Interfacial Tension
Composition
Equation 7-1. Dependence of the morphology of a ternary blend demonstrating complete wetting
on the interfacial tension of B and C and as a function of the composition of Phases B and C at
low shear rate. Note that in the schematics, yellow represents phase (A) which is the matrix, blue
represents phase (B) and green represents phase (C).
207

Interfacial Tension
Composition
Equation 7-2. Dependence of the morphology of a ternary blend demonstrating complete wetting
on the interfacial tension of B and C and as a function of the composition of Phases B and C at
high shear rate. Note that in the schematics, yellow represents phase (A) which is the matrix,
blue represents phase (B) and green represents phase (C).
208

CONCLUSION AND RECOMMENDATIONS
In this dissertation, we developed the phase morphology of multi-component blends via melt-
processing, assisted by solvent etching, in order to generate novel conductive devices with an
ultra-low percolation threshold. This thesis evaluates and develops continuity evolution in cocontinuous,
double-percolated, triple-percolated, and quadruple-percolated morphologies to find
a way to considerably reduce the percolation threshold of all phases in the blend.
In order to achieve the aforementioned objective (reducing the percolation threshold of the
conductive polymer and other phases), a comprehensive systematic study on binary, ternary,
quaternary, and quinary blends with particular complete wetting morphology is investigated for
the first time. In this case, reduction of the continuity and/or electrical percolation threshold of
phases, particularly of a conductive polymer, is carried out by developing the novel and original
structures defined as multi-percolated structures. These special structures are constituted of
hierarchically ordered self-assembled continuous phases in polymer blends. Since the first
condition for the preparation of hierarchically ordered phase structures in a multi-component
blend is satisfying all Harkins equation sets, based on the various interfacial tensions for the
polymer pairs in the blend, the components must be precisely selected to meet the condition. For
this reason, polyaniline and four other commercial polymers with specific surface tensions,
polarities and interfacial tensions are precisely selected. The interfacial tensions between the
various components range from 1 mN/m for PVDF/PMMA to 26.9 mN/m for PANI/HDPE.
Morphological sample characterization is performed by both SEM and FIB-AFM techniques.
The polymer blends are prepared in a melt-blending process in an internal mixer at 50 rpm for 8
min at 200°C. Selective solvent extraction of phases either assists in better detection of phases in
order to collect qualitative data, or is employed to produce porous samples. A combination of
FIB treatment and AFM technique allows us to clearly distinguish the phases based on the
topographical contrast formed by the difference in the FIB etching rate of materials. In order to
study the effect of viscosity ratio on the ternary blends, both low molecular weight and high
molecular weight poly(methyl methacrylate) are selected.
In this thesis, fundamental studies are developed as models to prepare novel morphologies such
as onion morphology and multi-percolated morphologies. By employing a melt-processable
209

conductive polymer, the applicability of these original morphologies is confirmed mostly by
conductivity results. It is shown that the order of phases in a complete wetting case such as a
multi-percolated structure is dependent only on the thermodynamic properties; although viscosity
and composition of phases play a major role in the determination of the morphology, they do not
change the order of phases. In this case, addition of the copolymer of adjacent phases in a multipercolated
structure induces a reduction in the phase size of the other phases, lowering the
percolation threshold of phases, which leads to higher conductivity of the samples. Though the
order of phases in multi-percolated structures is determined by the interfacial tension between
phases, it is shown that the viscosity of phases significantly influences the distribution of
interfacial energy between phases, and ultimately, the morphology.
It is shown that fully interconnected porous materials are prepared via the production of a multipercolated
structure followed by selective phase extraction. These porogens are used as
substrates for alternate deposition of PSS and PANI. A detailed study on the behaviour of
polyelectrolytes in the deposition process is carried out.
This thesis presents a systematic technique for study and prediction the morphology of
quaternary and quinary blends. To date, there is no theory for prediction of the morphology and
order of the phases in hierarchical self-assembled systems prepared by melt-blending. This is the
first study which begins to develop some mechanism and methods to predict the morphology of
multi-component blends.
210

REFERENCES
A. K. Gupta, K. R. S. (1993). Melt rheology and morphology of PP/SEBS/PC ternary blend.
Journal of Applied Polymer Science, 47(1), 167-184.
Adjari, A., Brochard-Wyart, F., de Gennes, P.-G., Leibler, L., Viovy, J.-L., & Rubinstein, M.
(1994). Slippage of an entangled polymer melt on a grafted surface. Physica A: Statistical
Mechanics and its Applications, 204(1-4), 17-39.
Al-Saleh, M. H., & Sundararaj, U. (2008). An innovative method to reduce percolation threshold
of carbon black filled immiscible polymer blends. Composites Part A: Applied Science
and Manufacturing, 39(Copyright 2008, The Institution of Engineering and Technology),
284-293.
Alexander, S. (1977). Polymer adsorption on small spheres. A scaling approach Journal de
Physique (Paris), 38(8), 977-981.
Anand, J., Palaniappan, S., & Sathyanarayana, D. N. (1998). Conducting polyaniline blends and
composites. Progress in Polymer Science (Oxford), 23(6), 993-1018.
Andreatta, A., Cao, Y., Chiang, J. C., Heeger, A. J., & Smith, P. (1988). Electrically-conductive
fibers of polyaniline spun from solutions in concentrated sulfuric acid. Synthetic Metals,
26(4), 383-389.
Angappane, S., Rajeev Kini, N., Natarajan, T. S., Rangarajan, G., & Wessling, B. (2002). PAni-
PMMA blend/metal Schottky barriers, Singapore, Singapore.
Antonoff, G. (1942). On the Validity of Antonoff's Rule. The Journal of Physical Chemistry,
46(4), 497-499.
Antonoff, G. N. (1907). J Chim Phys 5, 364-371.
Avgeropoulos, G. N., Weissert, F. C., Biddison, P. H., & Boehm, G. G. A. (1976). Hetrogeneous
Blends of Polymers Rheology and Morphology. Rubber Chemistry and Technology,
49(1), 93-104.
Baba, A., Kaneko, F., & Advincula, R. C. (1999). Adsorption properties of polyelectrolytes in
ultrathin multilayer assemblies: Investigations using the quartz crystal microbalance
(QCM) technique, Anaheim, CA, USA.
Baradie, B., & Shoichet, M. S. (2003). Insight into the Surface Properties of Fluorocarbon−Vinyl
Acetate Copolymer Films and Blends.. Macromolecules, 36(7), 2343-2348. doi:
10.1021/ma0216019
Bashforth, F., & Adams, J. C. (1883). An Attempt to Test the Theories of Capillary Action.
Cambridge: Cambridge Univ. Press and Deighton Bell & Co.
Bauhofer, W., & Kovacs, J. Z. (2009). A review and analysis of electrical percolation in carbon
nanotube polymer composites. Composites Science and Technology, 69(10), 1486-1498.
Berger, W., Kammer, H. W., & Kummerlowe, C. (1984). Makromol Chem Suppl B, 8, 101.
211

Bhattacharya, A., & De, A. (1999). Conducting polymers in solution - progress toward
processibility. Journal of Macromolecular Science - Reviews in Macromolecular
Chemistry and Physics, C39(1), 17-56.
Bhattacharya, S. (1986). Metal-Filled Polymers. New York: Marcel Dekker.
Bliznyuk, V. N., Baig, A., Singamaneni, S., Pud, A. A., Fatyeyeva, K. Y., & Shapoval, G. S.
(2005). Effects of surface and volume modification of poly(vinylidene fluoride) by
polyaniline on the structure and electrical properties of their composites. Polymer,
46(Copyright 2006, IEE), 11728-11736.
Blodgett, K. B. (1934). Monomolecular films of fatty acids on glass. J. Am. Chem. Soc., 56(2),
495-495.
Blunt, M., Barford, W., & Ball, R. (1989). Polymer adsorption and electron binding on rough
and fractal surfaces. Macromolecules, 22(3), 1458-1466.
Bourry, D., & Favis, B. D. (1998a). Cocontinuity and phase inversion in HDPE/PS blends:
Influence of interfacial modification and elasticity. Journal of Polymer Science, Part B:
Polymer Physics, 36(11), 1889-1899.
Braga, G. S., Paterno, L. G., Lima, J. P. H., Fonseca, F. J., & de Andrade, A. M. (2008).
Influence of the deposition parameters on the morphology and electrical conductivity of
PANI/PSS self-assembled films. Materials Science and Engineering C, 28(4), 555-562.
Bridge, B., & Tee, H. (1990). Large-scale simulation of the critical volume fraction for the
percolation threshold in metal-fibre-loaded polymer composites. International Journal of
Electronics, 69(6), 785-792.
Burroughes, J. H., Bradley, D. D. C., Brown, A. R., Marks, R. N., Mackay, K., Friend, R. H., et
al. (1990). Light-emitting diodes based on conjugated polymers. Nature, 347(6293), 539-
541.
Burroughes, J. H., Jones, C. A., & Friend, R. H. (1988). New Semiconductor Device Physics in
Polymer Diodes and Transistors. Nature, 335(6186), 137-141.
Calberg, C., Blacher, S., Gubbels, F., Brouers, F., Deltour, R., & Jerome, R. (1999). Electrical
and dielectric properties of carbon black filled co-continuous two-phase polymer blends.
Journal of Physics D: Applied Physics, 32(13), 1517-1525.
Campbell, D. J., Freidinger, E. R., Hastings, J. M., & Querns, M. K. (2002). Spontaneous
assembly of soda straws. Journal of Chemical Education, 79(2), 201.
Cao, Y., Smith, P., & Heeger, A. J. (1992). Counter-ion induced processibility of conducting
polyaniline and of conducting polyblends of polyaniline in bulk polymers. Synthetic
Metals, 48(1), 91-97.
Carmona, F. (1989). Conducting filled polymers. Physica A: Statistical and Theoretical Physics,
157(1), 461-469.
Carmona, F., Prudhon, P., & Barreau, F. (1984). Percolation in short fibres epoxy resin
composites: Conductivity behavior and finite size effects near threshold. Solid State
Communications, 51(4), 255-257.
212

Caruso, F. (2003). Hollow Inorganic Capsules via Colloid-Templated Layer-by-Layer
Electrostatic Assembly (pp. 145-168).
Caruso, F., Caruso, R. A., & Mohwald, H. (1998). Nanoengineering of inorganic and hybrid
hollow spheres by colloidal templating. Science, 282(Copyright 1999, IEE), 1111-1114.
Caruso, F., Caruso, R. A., & Mohwald, H. (1999). Production of Hollow Microspheres from
Nanostructured Composite Particles. Chemistry of Materials, 11(11), 3309-3314.
Caruso, F., Furlong, D. N., Ariga, K., Ichinose, I., & Kunitake, T. (1998). Characterization of
polyelectrolyte-protein multilayer films by atomic force microscopy, scanning electron
microscopy, and Fourier transform infrared reflection-absorption spectroscopy.
Langmuir, 14(16), 4559-4565.
Caruso, F., Lichtenfeld, H., Giersig, M., & Mohwald, H. (1998). Electrostatic Self-Assembly of
Silica Nanoparticle-Polyelectrolyte Multilayers on Polystyrene Latex Particles. J. Am.
Chem. Soc., 120(33), 8523-8524.
Caruso, F., & Schuler, C. (2000). Enzyme Multilayers on Colloid Particles: Assembly, Stability,
and Enzymatic Activity. Langmuir, 16(24), 9595-9603.
Cheung, J. H., Fou, A. F., Ferreira, M., & Rubner, M. F. (1993). Molecular self-assembly of
conducting polymers. A new layer-by-layer thin film deposition process. Polymer
Preprints, Division of Polymer Chemistry, American Chemical Society, 34(2), 757.
Cheung, J. H., Fou, A. F., & Rubner, M. F. (1994). Molecular self-assembly of conducting
polymers. Thin Solid Films, 244(1-2 pt 3), 985-989.
Cheung, J. H., Stockton, W. B., & Rubner, M. F. (1997). Molecular-level processing of
conjugated polymers. 3. Layer-by-layer manipulation of polyaniline via electrostatic
interactions. Macromolecules, 30(9), 2712-2716.
Chiang, C. K., Druy, M. A., Gau, S. C., Heeger, A. J., Louis, E. J., MacDiarmid, A. G., et al.
(1978). Synthesis of highly conducting films of derivatives of polyacetylene, (CH)x.
Journal of the American Chemical Society, 100(3), 1013-1015.
Chiang, C. K., Fincher, C. R., Park, Y. W., Heeger, A. J., Shirakawa, H., Louis, E. J., et al.
(1977). Electrical Conductivity in Doped Polyacetylene. Physical Review Letters, 39(17),
1098.
Chiang, C. K., Park, Y. W., Heeger, A. J., Shirakawa, H., Louis, E. J., & MacDiarmid, A. G.
(1978). Conducting polymers: Halogen doped polyacetylene. The Journal of Chemical
Physics, 69(11), 5098-5104.
Chipara, M. I., Munteanu, I., Notingher, P., Chipara, M. D., Reyes Romero, J., Negreanu, B., et
al. (1998). On polyaniline-polyvinylchloride blends, Vasteras, Sweden.
Chung, K. T., Sabo, A., & Pica, A. P. (1982). Electrical permittivity and conductivity of carbon
black-polyvinyl chloride composites. Journal of Applied Physics, 53(10), 6867-6879.
Clark, S. L., Montague, M. F., & Hammond, P. T. (1997). Ionic Effects of Sodium Chloride on
the Templated Deposition of Polyelectrolytes Using Layer-by-Layer Ionic Assembly.
Macromolecules, 30(23), 7237-7244.
Clausius, R. (1864). Abhandlungen über die mechanische wärme theorie, 2, 44.
213

Clerc, J. P., Giraud, G., Laugier, J. M., & Luck, J. M. (1990). The electrical conductivity of
binary disordered systems, percolation clusters, fractals and related models. Advances in
Physics, 39(3), 191 - 309.
Contractor, A. Q., Sureshkumar, T. N., Narayanan, R., Sukeerthi, S., Lal, R., & Srinivasa, R. S.
(1994). Conducting polymer-based biosensors. Electrochimica Acta, 39(8-9), 1321-1324.
Cox, W. P., & Merz, E. H. (1958). Correlation of dynamic and steady flow viscosities. Journal
of Polymer Science, 28(118), 619-622.
Dan, L., Yadong, J., Zhiming, W., Xiangdong, C., & Yanrong, L. (2000). Self-assembly of
polyaniline ultrathin films based on doping-induced deposition effect and applications for
chemical sensors, Beijing, China.
Dautzenberg, H., Jaeger, W., Kotz, J., Phillip, B., C., S., & Stscherbina, D. (1994).
Polyelectrolytes—Formation, Characterization, Application. Carl Hanser Verlag:
Munich, Germany.
Decher, G. (1997). Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites.
Science, 277(5330), 1232-1237.
Decher, G., & Hong, J. D. (1991a). Buildup of ultrathin multilayer films by a self-assembly
process, 1 consecutive adsorption of anionic and cationic bipolar amphiphiles on charged
surface. Chem.M acromol. Symp., 46, 321.
Decher, G., & Hong, J. D. (1991b). Buildup of ultrathin multilayer films by a self-assembly
process, 1. Consecutive adsorption of anionic and cationic bipolar amphiphiles on
charged surfaces. European Conference on Organized Organic Thin Films, 321.
Decher, G., Hong, J. D., & Schmitt, J. (1992a). Buildup of ultrathin multilayer films by a selfassembly
process: III. Consecutively alternating adsorption of anionic and cationic
polyelectrolytes on charged surfaces. [doi: DOI: 10.1016/0040-6090(92)90417-A]. Thin
Solid Films, 210-211(Part 2), 831-835.
Decher, G., Hong, J. D., & Schmitt, J. (1992b). Buildup of ultrathin multilayer films by a selfassembly
process: III. Consecutively alternating adsorption of anionic and cationic
polyelectrolytes on charged surfaces. Thin Solid Films, 210/211, 831-835.
Decher, G., Lvov, Y., & Schmitt, J. (1994). Proof of multilayer structural organization in selfassembled
polycation-polyanion molecular films. Thin Solid Films, 244(1-2 pt 3), 772-
777.
Decher, G., & Schlenoff, J. B. (2003). Multilayer Thin Films.
Decher, G., & Schmitt, J. (1992). Fine-tuning of the film thickness of ultrathin multilayer films
composed of consecutively alternating layers of anionic and cationic polyelectrolytes.
Progress in Colloid & Polymer Science, 89(Trends Colloid Interface Sci. VI), 160-164.
DeLongchamp, D., & Hammond, P. T. (2001). Layer-by-layer assembly of PEDOT/polyaniline
electrochromic devices. Advanced Materials, 13(19), 1455-1459.
Denbigh, K. G. (1971). The principles of chemical equilibrium, with applications in chemistry
and chemical engineering. Cambridge [Eng.: University Press.
214

Domenech, S. C., Bortoluzzi, J. H., Soldi, V., & Franco, C. V. (2002). Electroactive mixtures of
polyaniline and EPDM rubber: Chemical, thermal, and morphological characterization.
Journal of Applied Polymer Science, 87(3), 535-547.
Donahue, D. J., & Bartell, F. E. (1952). J. Phys. Chem., 56, 480.
Dong-Uk, C., Kitae, S., Young Chul, K., Yun Heum, P., & Jun Young, L. (2001). Electrically
conducting polymer blends with thermally processable polyaniline, Kyoto, Japan.
Du, F., Scogna, R. C., Zhou, W., Brand, S., Fischer, J. E., & Winey, K. I. (2004). Nanotube
Networks in Polymer Nanocomposites: Rheology and Electrical Conductivity.
Macromolecules, 37(24), 9048-9055.
Dubas, S. T., & Schlenoff, J. B. (1999). Factors Controlling the Growth of Polyelectrolyte
Multilayers. Macromolecules, 32(24), 8153-8160.
Dubas, S. T., & Schlenoff, J. B. (2001b). Polyelectrolyte Multilayers Containing a Weak
Polyacid:&nbsp; Construction and Deconstruction. Macromolecules, 34(11), 3736-3740.
Dubas, S. T., & Schlenoff, J. B. (2001c). Swelling and Smoothing of Polyelectrolyte Multilayers
by Salt. Langmuir, 17(25), 7725-7727. doi: doi:10.1021/la0112099
Epstein, A. J., Ginder, J. M., Zuo, F., Woo, H. S., Tanner, D. B., Richter, A. F., et al. (1986).
Insulator-to-metal transition in polyaniline: effect of protonation in emeraldine. Synthetic
Metals, 21(1), 63-70.
Everaert, V., Aerts, L., & Groeninckx, G. (1999). Phase morphology development in immiscible
PP/(PS/PPE) blends influence of the melt-viscosity ratio and blend composition.
Polymer, 40(24), 6627-6644.
Farhat, T., Yassin, G., Dubas, S. T., & Schlenoff, J. B. (1999). Water and ion pairing in
polyelectrolyte multilayers. Langmuir, 15(20), 6621-6623.
Farhat, T. R., & Schlenoff, J. B. (2002). Corrosion control using polyelectrolyte multilayers.
Electrochemical and Solid-State Letters, 5(4), 13-15.
Favis, B. D. (2000). Factors Influencing the Morphology in Immiscible Polymer Blends in Melt
Processing. In D. R. Paul & C. B. Bucknall (Eds.), Polymer Blends: Formulation and
Performance, Two-Volume Set (Vol. 1, pp. 239-289). New York: John Wiley & Sons,
Inc.
Favis, B. D., & Chalifoux, J. P. (1988). Influence of composition on the morphology of
polypropylene/polycarbonate blends. Polymer, 29(10), 1761-1767.
Feng, L., & Kamal, M. R. (2005). Crystallization and melting behavior of homogeneous and
heterogeneous linear low-density polyethylene resins. Polymer Engineering and Science,
45(Copyright 2006, IEE), 1140-1151.
Ferreira, M., Cheung, J. H., & Rubner, M. F. (1994). Molecular self-assembly of conjugated
polyions: A new process for fabricating multilayer thin film heterostructures. Thin Solid
Films, 244(1-2 pt 3), 806-809.
Ferreira, M., & Rubner, M. F. (1995). Molecular-level processing of conjugated polymers. 1.
Layer-by-layer manipulation of conjugated polyions. Macromolecules, 28(21), 7107-
7114.
215

Flandin, L., Bréchet, Y., & Cavaillé, J. Y. (2001). Electrically conductive polymer
nanocomposites as deformation sensors. Composites Science and Technology, 61(6), 895-
901.
Flory, P. J. (1942). Thermodynamics of High Polymer Solutions. The Journal of Chemical
Physics, 10(1), 51-61.
Flory, P. J. (1944). Thermodynamics of Heterogeneous Polymers and Their Solutions. The
Journal of Chemical Physics, 12(11), 425-438.
Flory, P. J. (1953). Principles of Polymer Chemistry. New York: Cornell Univ. Press.
Flory, P. J. (1965). Statistical Thermodynamics of Liquid Mixtures. Journal of the American
Chemical Society, 87(9), 1833-1838.
Flory, P. J., Orwoll, R. A., & Vrij, A. (1964). Statistical Thermodynamics of Chain Molecule
Liquids. I. An Equation of State for Normal Paraffin Hydrocarbons. Journal of the
American Chemical Society, 86(17), 3507-3514.
Fou, A. C., Onitsuka, O., Ferreira, M., & Rubner, M. F. (1996). Fabrication and properties of
light-emitting diodes based on self-assembled multilayers of poly(phenylene vinylene).
Journal of Applied Physics, 79(10), 7501-7509.
Fou, A. C., & Rubner, M. F. (2002). Molecular-Level Processing of Conjugated Polymers. 2.
Layer-by-Layer Manipulation of In-Situ Polymerized p-Type Doped Conducting
Polymers. Macromolecules, 28(21), 7115-7120.
Foulger, S. H. (1999). Reduced percolation thresholds of immiscible conductive blends. Journal
of Polymer Science Part B: Polymer Physics, 37(15), 1899-1910.
Fraysse, J., Planes, J., & Dufresne, A. (2000). Transport and mechanical properties of
polyaniline - Poly(methyl methacrylate) blends exhibiting low percolation threshold,
Poznan.
Fricke, H. A. (1924). mathematical treatment of the electrical conductivity and capacity of
disperse systems. I. The electrical conductivity of a suspension of homogeneous
spheroids. J. Physical Rev.,, 24, 575-587.
Galal, A., Atta, N. F., & Mark Jr, H. B. (1998). Sensors Based on Organic Conducting-Polymer
Electrodes. ACS Symposium Series, 690, 210-230.
Garnier, F., Hajlaoui, R., Yassar, A., & Srivastava, P. (1994). All-polymer field-effect transistor
realized by printing techniques. Science, 265(Copyright 1994, IEE), 1684-1686.
Gaylord, N. G. (1969). USA Patent No.: U. Patent.
Gaylord, N. G., ,Vol.142,P.76,(ACS,Washington,1975). (1975). Copolymers, Polyblends and
Composites. ACS Advances in Chemistry Series, 142, 76.
Gelves, G. A., Lin, B., Sundararaj, U., & Haber, J. A. (2006). Low electrical percolation
threshold of silver and copper nanowires in polystyrene composites. Advanced
Functional Materials, 16(Compendex), 2423-2430.
Gerard, M., Chaubey, A., & Malhotra, B. D. (2002). Application of conducting polymers to
biosensors. Biosensors and Bioelectronics, 17(5), 345-359.
216

Geuskens, G., Gielens, J. L., Geshef, D., & Deltour, R. (1987). The electrical conductivity of
polymer blends filled with carbon-black. European Polymer Journal, 23(12), 993-995.
Gibbs, J. W. (1876). Trans. Conn. Acad. Sci., 3228.
Ginder, J. M., Richter, A. F., MacDiarmid, A. G., & Epstein, A. J. (1987). Insulator-to-metal
Transition in Polyaniline. Solid State Communications, 63(2), 97-101.
Girifalco, L. A., & Good, R. J. (1957). A Theory for the Estimation of Surface and Interfacial
Energies. I. Derivation and Application to Interfacial Tension. The Journal of Physical
Chemistry, 61(7), 904-909.
Goh, S. H., Chan, H. S. O., & Ong, C. H. (1998). Miscible blends of conductive polyaniline with
tertiary amide polymers. Journal of Applied Polymer Science, 68(11), 1839-1844.
Golestanian, R., & Sens, P. (1999). Comment on &ldquo;Adsorption of polyelectrolyte onto a
colloid of opposite charge&rdquo; [and reply]. Physical Review Letters, 83(12), 2473-
2474.
Good, R. J., & Girifalco, L. A. (1964). Adv. Chem., 43, 74.
Graul, T. W., & Schlenoff, J. B. (1999). Capillaries modified by polyelectrolyte multilayers for
electrophoretic separations. Analytical Chemistry, 71(18), 4007-4013.
Greene, R. L., Street, G. B., & Suter, L. J. (1975). Superconductivity in Polysulfur Nitride
(SN)_{X}. Physical Review Letters, 34(10), 577.
Grimmett, G. R. (1999). Percolation: Springer.
Gubbels, F., Blacher, S., Vanlathem, E., Jerome, R., Deltour, R., Brouers, F., et al. (1995).
Design of electrical conductive composites: key role of the morphology on the electrical
properties of carbon black filled polymer blends. Macromolecules, 28(5), 1559-1566.
Gubbels, F., Jerome, R., Teyssie, P., Vanlathem, E., Deltour, R., Calderone, A., et al. (1994).
Selective localization of carbon black in immiscible polymer blends: a useful tool to
design electrical conductive composites. Macromolecules, 27(7), 1972-1974.
Guo, H. F., Gvozdic, N. V., & Meier, D. J. (1997). Prediction and manipulation of the phase
morphologies of multiphase polymer blends: II. Quaternary systems. Polymer, 38(19),
4915-4923.
Guo, H. F., Packirisamy, S., Gvozdic, N. V., & Meier, D. J. (1997). Prediction and manipulation
of the phase morphologies of multiphase polymer blends: 1. Ternary systems. Polymer,
38(4), 785-794.
Gurovitch, E., & Sens, P. (1999). Adsorption of Polyelectrolyte onto a Colloid of Opposite
Charge. Physical Review Letters, 82(2), 339-342.
Ha, B. Y., & Thirumalai, D. (1995). Electrostatic persistence length of a polyelectrolyte chain.
Macromolecules, 28(Compendex), 577-581.
Haba, Y., Segal, E., Narkis, M., Titelman, G. I., & Siegmann, A. (2000). Polyaniline-
DBSA/polymer blends prepared via aqueous dispersions. Synthetic Metals, 110(3), 189-
193.
217

Halls, J. J. M., Walsh, C. A., Greenham, N. C., Marseglia, E. A., Friend, R. H., Moratti, S. C., et
al. (1995). Efficient photodiodes from interpenetrating polymer networks. Nature,
376(6540), 498-500.
Hammond, P. T. (2004). Form and Function in Multilayer Assembly: New Applications at the
Nanoscale. Advanced Materials, 16(15), 1271-1293.
Han, M. G., & Im, S. S. (1999). Electrical and structural analysis of conductive
polyaniline/polyimide blends. Journal of Applied Polymer Science, 71(13), 2169-2178.
Hara, M., & Sauer, J. A. (1998). Synergism in mechanical properties of polymer/polymer blends.
Journal of Macromolecular Science - Reviews in Macromolecular Chemistry and
Physics, C38(2), 327-362.
Harkins, W. D., & Feldman, A. (1922). Films. The Spreading of Liquids and the Spreading
Coefficient. Journal of the American Chemical Society, 44(12), 2665-2685. doi:
10.1021/ja01433a001
Harrats, C., Omonov, T. S., & Groeninckx, G. (2005). Co-continuous and encapsulated three
phase morphologies in uncompatibilized and reactively compatibilized polyamide
6/polypropylene/polystyrene ternary blends using two reactive precursors. Polymer,
46(Copyright 2006, IEE), 12322-12336.
Harris, J. J., & Bruening, M. L. (2000). Electrochemical and in Situ Ellipsometric Investigation
of the Permeability and Stability of Layered Polyelectrolyte Films. Langmuir, 16(4),
2006-2013.
Hartikainen, J., Lahtinen, M., Torkkeli, M., Serimaa, R., Valkonen, J., Rissanen, K., et al.
(2001). Comb-Shaped Supramolecules Based on Protonated Polyaniline and Their Self-
Organization into Nanoscale Structures: Polyaniline Sulfonates/Zinc Sulfonates.
Macromolecules, 34(22), 7789-7795.
He, J., Bu, W., & Zeng, J. (1997). Co-phase continuity in immiscible binary polymer blends.
Polymer, 38(26), 6347-6353.
Heywang, G., & Jonas, F. (1992). Poly(alkylenedioxythiophene)s - new, very stable conducting
polymers. Advanced Materials, 4(2), 116-118.
Ho, R. M., Wu, C. H., & Su, A. C. (1990). Morphology of plastic/rubber blends. Polymer
Engineering and Science, 30(9), 511-518.
Hobbs, S. Y., Dekkers, M. E. J., & Watkins, V. H. (1988). Effect of interfacial forces on polymer
blend morphologies. Polymer, 29(9), 1598-1602.
Hoganand, J. P., & Banks, R. L. (1958). US Patent.
Hone, D., Ji, H., & Pincus, P. A. (1987). Polymer adsorption on rough surfaces. 1. Ideal long
chain. Macromolecules, 20(10), 2543-2549.
Horiuchi, S., Matchariyakul, N., Yase, K., & Kitano, T. (1997). Morphology Development
through an Interfacial Reaction in Ternary Immiscible Polymer Blends. Macromolecules,
30(12), 3664-3670.
Hosokawa, K., Shimoyama, I., & Miura, H. (1996). Two-dimensional micro-self-assembly using
the surface tension of water. Sensors and Actuators, A: Physical, 57(2), 117-125.
218

Hou, S., Harrell, C. C., Trofin, L., Kohli, P., & Martin, C. R. (2004). Layer-by-Layer Nanotube
Template Synthesis. Journal of the American Chemical Society, 126(18), 5674-5675.
http://web.mit.edu/newsoffice/2009/cartilage-0511.html.
http://www.chem.fsu.edu/multilayers/.
http://www.veeco.com/library/resources/Sample_Preparation/index.aspx.
Huang, W. S., D., H. B., & MacDiarmid, A. G. (1986). J. Chem. Soc. Faraday Trans., 82, 2385.
Huggins, M. L. (1942). Some Properties of Solutions of Long-chain Compounds. The Journal of
Physical Chemistry, 46(1), 151-158.
Ikkala, O. T., Laakso, J., Väkiparta, K., Virtanen, E., Ruohonen, H., Järvinen, H., et al. (1995).
Counter-ion induced processibility of polyaniline: Conducting melt processible polymer
blends. Synthetic Metals, 69(1-3), 97-100.
Inzelt, G., Pineri, M., Schultze, J. W., & Vorotyntsev, M. A. (2000). Electron and proton
conducting polymers: Recent developments and prospects. Electrochimica Acta, 45(15-
16), 2403-2421.
Israelachvili, J. (1992). Intermolecular and Surface Forces, Second Edition: With Applications to
Colloidal and Biological Systems (Colloid Science): {Academic Press}.
Ito, T., Shirakawa, H., & Ikeda, S. (1996). Simultaneous polymerization and formation of
polyacetylene film on the surface of concentrated soluble Ziegler-type catalyst solution.
Journal of Polymer Science, Part A: Polymer Chemistry, 34(13), 2533-2542.
Jaber, J. A., & Schlenoff, J. B. (2007). Counterfoils and water in polyelectrolyte multilayers: A
tale of two polycations. Langmuir, 23(2), 896-901.
Jaffar, S., Nam, K. T., Khademhosseini, A., Xing, J., Langer, R., & Belcher, A. (2004). Layerby-layer
surface modification and patterned electrostatic deposition of quantum dots.
Nano Letters, 4(8), 1421-1425.
Jagur-Grodzinski, J. (2002). Electronically conductive polymers. Polymers for Advanced
Technologies, 13(9), 615-625.
Jing, X., Yang, Q., Zheng, M., & Lan, L. (1999). Study on the electrically conductive
polyamide/polyaniline blends. Hsi-An Chiao Tung Ta Hsueh/Journal of Xi'an Jiaotong
University, 33(8), 74-77.
Joanny, J. F. (1999). Polyelectrolyte adsorption and charge inversion. The European Physical
Journal B - Condensed Matter and Complex Systems, V9(1), 117-122.
Johnson, K. L., Kendall, K., & Roberts, A. D. (1971). Surface Energy and the Contact of Elastic
Solids. Proceedings of the Royal Society of London. A. Mathematical and Physical
Sciences, 324(1558), 301-313.
Joly, S., Kane, R., Radzilowski, L., Wang, T., Wu, A., Cohen, R. E., et al. (2000). Multilayer
nanoreactors for metallic and semiconducting particles. Langmuir, 16(3), 1354-1359.
Jordhamo, G. M., Manson, J. A., & Sperling, L. H. (1986). Phase Continuity and Inversion in
Polymer Blends and Simultaneous Interpenetrating Networks. Polymer Engineering and
Science, 26(8), 517-524.
219

Jousseaume, V., Morsli, A., & Bonnet, A. (2002). Thermal behavior of polyaniline films and
polyaniline-polystyrene blends. Journal of Applied Polymer Science, 84(10), 1848-1855.
Jousseaume, V., Morsli, M., Bonnet, A., Tesson, O., & Lefrant, S. (1998). Electrical properties
of polyaniline-polystyrene blends above the percolation threshold. Journal of Applied
Polymer Science, 67(7), 1205-1208.
Kababya, S., Appel, M., Haba, Y., Titelman, G. I., & Schmidt, A. (1999).
Polyaniline−Dodecylbenzene Sulfonic Acid Polymerized from Aqueous Medium: A
Solid State NMR Characterization. Macromolecules, 32(16), 5357-5364.
Kaneto, K., Kaneko, M., Min, Y., & MacDiarmid, A. G. (1995). "Artificial muscle":
Electromechanical actuators using polyaniline films. Synthetic Metals, 71(1-3), 2211-
2212.
Keller, S. W., Kim, H.-N., & Mallouk, T. E. (1994). Layer-by-Layer Assembly of Intercalation
Compounds and Heterostructures on Surfaces: Toward Molecular "Beaker" Epitaxy. J.
Am. Chem. Soc., 116(19), 8817-8818.
Kim, H. S., Sohn, B. H., Lee, W., Lee, J. K., Choi, S. J., & Kwon, S. J. (2002). Multifunctional
layer-by-layer self-assembly of conducting polymers and magnetic nanoparticles. Thin
Solid Films, 419(1-2), 173-177.
Kirkpatrick, S. (1973). Percolation and Conduction. Reviews of Modern Physics, 45(Copyright
(C) 2009 The American Physical Society), 574.
Klason, C., & Kubát, J. (1975). Anomalous behavior of electrical conductivity and thermal noise
in carbon black-containing polymers at Tg and Tm. Journal of Applied Polymer Science,
19(3), 831-845.
Klitzing, R., & Moehwald, H. (1995). Proton Concentration Profile in Ultrathin Polyelectrolyte
Films. Langmuir, 11(9), 3554-3559.
Klitzing, R. v., Espert, A., Asnacios, A., Hellweg, T., Colin, A., & Langevin, D. (1999). Forces
in foam films containing polyelectrolyte and surfactant. Colloids and Surfaces A:
Physicochemical and Engineering Aspects, 149(1-3), 131-140.
Koh, J. C. Y., & Fortini, A. (1973). Prediction of thermal conductivity and electrical resistivity
of porous metallic materials. Int. J. Heat Mass Transfer,, 16, 2013-2021.
Kotov, N. A., Dekany, I., & Fendler, J. H. (1995). Layer-by-Layer Self-Assembly of
Polyelectrolyte-Semiconductor Nanoparticle Composite Films. J. Phys. Chem., 99(35),
13065-13069.
Kotov, N. A., Haraszti, T., Turi, L., Zavala, G., Geer, R. E., Dekany, I., et al. (1997). Mechanism
of and defect formation in the self-assembly of polymeric polycation-montmorillonite
ultrathin films. Journal of the American Chemical Society, 119(29), 6821-6832.
Kraft, A., Grimsdale, Andrew C., & Holmes, Andrew B. (1998). Elektrolumineszierende
konjugierte Polymere - Polymere erstrahlen in neuem Licht. Angewandte Chemie, 110(4),
416-443.
220

Krass, H., Papastavrou, G., & Kurth, D. G. (2003). Layer-by-layer self-assembly of a
polyelectrolyte bearing metal ion coordination and electrostatic functionality. Chemistry
of Materials, 15(1), 196-203.
Krieger, I. M., & Dougherty, T. J. (1959). A Mechanism for Non-Newtonian Flow in
Suspensions of Rigid Spheres. Journal of Rheology, 3(1), 137-152.
Kumar, D., & Sharma, R. C. (1998). Advances in conductive polymers. European Polymer
Journal, 34(8), 1053-1060.
Kumin, Y., & Chang Dae, H. (1996). Effects of shear flow and annealing on the morphology of
rapidly precipitated immiscible blends of polystyrene and polyisoprene. Polymer, 37(26),
5795-5805.
Kunze, K. K., & Netz, R. R. (2002). Morphologies of semiflexible polyelectrolyte complexes.
Europhysics Letters, 58(2), 299-305.
Ladam, G., Schaad, P., Voegel, J. C., Schaaf, P., Decher, G., & Cuisinier, F. (2000). In situ
determination of the structural properties of initially deposited polyelectrolyte
multilayers. Langmuir, 16(3), 1249-1255.
Laska, J., Zak, K., & Pron, A. (1997). Conducting blends of polyaniline with conventional
polymers. Synthetic Metals, 84(1-3 pt 1), 117-118.
Lavalle, P., Gergely, C., Cuisinier, F. J. G., Decher, G., Schaaf, P., Voegel, J. C., et al. (2002).
Comparison of the structure of polyelectrolyte multilayer films exhibiting a linear and an
exponential growth regime: An in situ atomic force microscopy study. Macromolecules,
35(11), 4458-4465.
Lavalle, P., Picart, C., Mutterer, J., Gergely, C., Reiss, H., Voegel, J.-C., et al. (2004). Modeling
the Buildup of Polyelectrolyte Multilayer Films Having Exponential Growth. Journal of
Physical Chemistry B, 108(2), 635-648.
Lefaux, C. J., Zimberlin, J. A., Dobrynin, A. V., & Mather, P. T. (2004). Polyelectrolyte spin
assembly: influence of ionic strength on the growth of multilayered thin films. Journal of
Polymer Science, Part B (Polymer Physics), 42(19), 3654-3666.
Legros, A., Carreau, P. J., Favis, B. D., & Michel, A. (1997). Morphology modification by
interfacial chemical reaction in a polyester/ethylene vinyl acetate/polyethylene blend.
Polymer, 38(20), 5085-5089.
Levon, K., Margolina, A., & Patashinsky, A. Z. (1993). Multiple percolation in conducting
polymer blends. Macromolecules, 26(15), 4061-4063.
Levon, K., Margolina, A., & Patashinsky, A. Z. (2002). Multiple percolation in conducting
polymer blends. Macromolecules, 26(15), 4061-4063.
Li, J., & Favis, B. D. (2001a). Characterizing co-continuous high density
polyethylene/polystyrene blends. Polymer, 42(11), 5047-5053.
Li, J., Ma, P. L., & Favis, B. D. (2002). The role of the blend interface type on morphology in
cocontinuous polymer blends. Macromolecules, 35(6), 2005-2016.
221

Liang, Z., Cabarcos, O. M., Allara, D. L., & Wang, Q. (2004). Hydrogen-Bonding-Directed
Layer-by-Layer Assembly of Conjugated Polymers. Advanced Materials, 16(9-10), 823-
827.
Liangcai, L., Ming, W., Huoming, S., Haiying, L., Qingdong, Q., & Yuanlong, D. (2001).
Preparation and EIS studies on polyimide/polyaniline blend film for corrosion protection.
Polymers for Advanced Technologies, 12(11-12), 720-723.
Lifshitz, E. M., Landau, L. D., & Pitaevskii, L. P. (1984). Electrodynamics of Continuous Media
(Vol. 8).
Lin, B., Gelves, G. A., Haber, J. A., Potschke, P., & Sundararaj, U. (2008). Electrical,
morphological and rheological study of melt-mixed polystyrene/copper nanowire
nanocomposites. Macromolecular Materials and Engineering, 293(Compendex), 631-
640.
Love, J. C., Estroff, L. A., Kriebel, J. K., Nuzzo, R. G., & Whitesides, G. M. (2005). Self-
Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chemical
Reviews, 105(4), 1103-1170.
Luciani, A., & Jarrin, J. (1996). Morphology development in immiscible polymer blends.
Polymer Engineering & Science, 36(12), 1619-1626.
Luckham, P. F., & Klein, J. (1985). Interactions between smooth solid surfaces in solutions of
adsorbing and nonadsorbing polymers in good solvent conditions. Macromolecules,
18(4), 721-728.
Lutt, M., Fitzsimmons, M. R., & DeQuan, L. (1998). X-ray reflectivity study of self-assembled
thin films of macrocycles and macromolecules. Journal of Physical Chemistry B, 102(2),
400-405.
Luzinov, I., Pagnoulle, C., & Jerome, R. (2000). Ternary polymer blend with core-shell
dispersed phases: effect of the core-forming polymer on phase morphology and
mechanical properties. Polymer, 41(19), 7099-7109.
Lvov, Y., Ariga, K., Ichinose, I., & Kunitake, T. (1995). Assembly of Multicomponent Protein
Films by Means of Electrostatic Layer-by-Layer Adsorption. J. Am. Chem. Soc., 117(22),
6117-6123.
Lvov, Y., Decher, G., & Sukhorukov, G. (1993). Assembly of thin films by means of successive
deposition of alternate layers of DNA and poly(allylamine). Macromolecules, 26(20),
5396-5399.
Lvov, Y. M., & Decher, G. (1994). Assembly of multilayer ordered films by alternating
adsorption of oppositely charged macromolecules. Crystallography Reports, 39(4), 628-
647.
Lyngaae-Jorgensen, J., & Utracki, L. A. (2003). Structuring polymer blends with bicontinuous
phase morphology. II. Tailoring blends with ultralow critical volume fraction. Polymer,
44(5), 1661-1669.
M. Dimitrova, Y. A. P. L. F. M. M. W. C. S. Y. H. J. C. V. J. O. (2007). Adenoviral Gene
Delivery from Multilayered Polyelectrolyte Architectures. Advanced Functional
Materials, 17(2), 233-245.
222

Macdiarmid, A. G., Jin-Chih, C., Halpern, M., Wu-Song, H., Shao-Lin, M., Somasiri, L. D., et
al. (1985). `Polyaniline': interconversion of metallic and insulating forms. Paper
presented at the Proceedings of the International Conference on the Physics and
Chemistry of Low-Dimensional Synthetic Metals (ICSM 84), 17-22 June 1984, UK.
Malek, G., Johnson, L. V., Mace, B. E., Saloupis, P., Schmechel, D. E., Rickman, D. W., et al.
(2005). Apolipoprotein E allele-dependent pathogenesis: A model for age-related retinal
degeneration. Proceedings of the National Academy of Sciences of the United States of
America, 102(33), 11900-11905.
Malmonge, L. F., Lopes, G. d. A., Langiano, S. d. C., Malmonge, J. A., Cordeiro, J. M. M., &
Mattoso, L. H. C. (2006). A new route to obtain PVDF/PANI conducting blends.
European Polymer Journal, 42(11), 3108-3113.
Man, W., & Shaw, L. (2006). Electrical and mechanical behaviors of carbon nanotube- filled
polymer blends. Journal of Applied Polymer Science, 99(2), 477-488.
Maoz, R., Frydman, E., Cohen, S. R., & Sagiv, J. (2000). Constructive nanolithography: Sitedefined
silver self-assembly on nanoelectrochemically patterned monolayer templates.
Advanced Materials, 12(6), 424-429.
Margolis, J. M. (1989). Conductive Polymers and Plastics. New York: Chapman and Hall.
Mateescu, E. M., Jeppesen, C., & Pincus, P. (1999). Overcharging of a spherical macroion by an
oppositely charged polyelectrolyte. Europhysics Letters, 46(4), 493-498.
Mathur, P., Misra, S. C. K., Singh, B. P., & Tripathi, P. (2002). Semiconducting polyaniline
nanocrystalline thin films for electronic and sensor applications, Delhi, India.
Matos, M., Favis, B. D., & Lomellini, P. (1995). Interfacial modification of polymer blends--the
emulsification curve: 1. Influence of molecular weight and chemical composition of the
interfacial modifier. Polymer, 36(20), 3899-3907.
Mattoso, L. H. C., Leite, F. L., Paterno, L. G., Borato, C. E., Herrmann, P. S. P., & Oliveira, O.
N., Jr. (2005). Study on the adsorption of poly(o-ethoxyaniline) nanostructured films
using atomic force microscopy. Polymer, 46(26), 12503-12510.
McMaster, L. P. (2002). Aspects of Polymer-Polymer Thermodynamics. [doi:
10.1021/ma60035a024]. Macromolecules, 6(5), 760-773.
Mekhilef, N., & Verhoogt, H. (1996). Phase inversion and dual-phase continuity in polymer
blends: theoretical predictions and experimental results. Polymer, 37(18), 4069-4077.
Meli, M.-V., & Lennox, R. B. (2003). Preparation of Nanoscale Au Islands in Patterned Arrays.
Langmuir, 19(22), 9097-9100.
Metelkin, V. I., & Blekht, V. S. (1984). Formation of a Continuous Phase in Heterogeneous
Polymer Mixtures. Colloid Journal of the USSR (English Translation of Kolloidnyi
Zhurnal), 46(3), 425-429.
Miles, I. S., & Zurek, A. (1988a). Preparation, structure, and properties of two-phase cocontinuous
polymer blends. Polymer Engineering & Science, 28(12), 796-805.
223

Miyasaka, K., Watanabe, K., Jojima, E., Aida, H., Sumita, M., & Ishikawa, K. (1982a).
Electrical conductivity of carbon-polymer composites as a function of carbon content.
Journal of Materials Science, 17(6), 1610-1616.
Molau, G. E., & Keskkula, H. (1966). Heterogeneous polymer systems. IV. Mechanism of
rubber particle formation in rubber-modified vinyl polymers. Journal of Polymer Science
Part A-1: Polymer Chemistry, 4(6), 1595-1607.
Montes, J. M., Cuevas, F. G., Rodriguez, J. A., & Herrera, E. J. (2005). Electrical conductivity of
sintered powder compacts. Powder Metallurgy, 48(4), 343-344.
Moon Gyu, H., & Seung Soon, I. (2001). Morphological study of conductive
polyaniline/polyimide blends. I. Determination of compatibility by small-angle X-ray
scattering method. Polymer, 42(17), 7449-7454.
Morgan, H., Foot, P. J. S., & Brooks, N. W. (2001). The effects of composition and processing
variables on the properties of thermoplastic polyaniline blends and composites. Journal
of Materials Science, 36(22), 5369-5377.
Moussaif, N., & Jérôme, R. (1999). Compatibilization of immiscible polymer blends (PC/PVDF)
by the addition of a third polymer (PMMA): analysis of phase morphology and
mechanical properties. Polymer, 40(14), 3919-3932.
Munson-McGee, S. H. (1991). Estimation of the critical concentration in an anisotropic
percolation network. Physical Review B (Condensed Matter), 43(4), 3331-3336.
Narkis, M., Haba, Y., Segal, E., Zilberman, M., Titelman, G. I., & Siegmann, A. (2000a).
Structured electrically conductive polyaniline/polymer blends. Polymers for Advanced
Technologies, 11(8-12), 665-673.
Narkis, M., & Vaxman, A. (1984). Resistivity behavior of filled electrically conductive
crosslinked polyethylene. Journal of Applied Polymer Science, 29(5), 1639-1652.
Narkis, M., Zilberman, M., & Siegmann, A. (1997). On the &quot;curiosity&quot; of electrically
conductive melt processed doped-polyaniline/polymer blends versus carbonblack/polymer
compounds. Polymers for Advanced Technologies, 8(8), 525-528.
Nauman, E. B., & He, D. Q. (2001). Nonlinear diffusion and phase separation. Chemical
Engineering Science, 56(6), 1999-2018.
Nemirovski, N., Siegmann, A., & Narkis, M. (1995). Morphology of ternary immiscible polymer
blends. Journal of Macromolecular Science - Physics, B34(4), 459-475.
Netz, R. R., & Andelman, D. (2003). Neutral and charged polymers at interfaces. Physics
Reports, 380(1-2), 1-95.
Nguyen, T. T., & Shklovskii, B. I. (2001). Overcharging of a macroion by an oppositely charged
polyelectrolyte. Physica A, 293(3-4), 324-338.
Nicol, E., Habib-Jiwan, J. L., & Jonas, A. M. (2003). Polyelectrolyte Multilayers as
Nanocontainers for Functional Hydrophilic Molecules. Langmuir, 19(15), 6178-6186.
Nicolau, Y. F. (1985). Solution deposition of thin solid compound films by a successive ioniclayer
adsorption and reaction process, Sydney, NSW, Australia.
224

Niziol, J., & Laska, J. (1999). Conductivity of blends of polyaniline with PMMA and cellulose
acetate: aging studies, Montpellier, France.
Odijk, T. (1980). Binding of Long Flexible Chains to a Rodlike Macromolecule.
Macromolecules, 13(6), 1542-1546.
Omonov, T. S., Harrats, C., & Groeninckx, G. (2005). Co-continuous and encapsulated three
phase morphologies in uncompatibilized and reactively compatibilized polyamide
6/polypropylene/polystyrene ternary blends using two reactive precursors. Polymer,
46(26), 12322-12336.
Ong, C. H., Goh, S. H., & Chan, H. S. O. (1997). Conductive polyaniline/poly(vinylphosphonic
acid) blends. Polymer Bulletin (Berlin), 39(5), 627.
P. Bertrand, A. J. A. L. R. L. (2000). Ultrathin polymer coatings by complexation of
polyelectrolytes at interfaces: suitable materials, structure and properties.
Macromolecular Rapid Communications, 21(7), 319-348.
Paloheimo, J., Laakso, K., Isotalo, H., & Stubb, H. (1995). Conductivity, thermoelectric power
and field-effect mobility in self-assembled films of polyanilines and oligoanilines.
Synthetic Metals, 68(3), 249-257.
Panipol. (2000). Polyaniline - Processing and Applications.
http://www.azom.com/details.asp?ArticleID=1197.
Park, M.-K., Deng, S., & Advincula, R. C. (2005). Sustained Release Control via Photo-Cross-
Linking of Polyelectrolyte Layer-by-Layer Hollow Capsules. Langmuir, 21(12), 5272-
5277.
Paul, D., & Newman, S. (1978). Polymer Blends (Vol. 1, 2). New York: Academic Press.
Paul, D. R., & Barlow, J. W. (1980b). Polymer Blends (or Alloys). Journal of Macromolecular
Science- Reviews in Macromolecular Chemistry, C18(1), 109-168.
Paul, R. K., Vijayanathan, V., & Pillai, C. K. S. (1999). Melt/solution processable conducting
polyaniline: doping studies with a novel phosphoric acid ester. Synthetic Metals, 104(3),
189-195.
Picart, C., Ladam, G., Senger, B., Voegel, J. C., Schaaf, P., Cuisinier, F. J. G., et al. (2001).
Determination of structural parameters characterizing thin films by optical methods: A
comparison between scanning angle reflectometry and optical waveguide lightmode
spectroscopy. Journal of Chemical Physics, 115(2), 1086-1094.
Picart, C., Lavalle, P., Hubert, P., Cuisinier, F. J. G., Decher, G., Schaaf, P., et al. (2001).
Buildup mechanism for poly(L-lysine)/hyaluronic acid films onto a solid surface.
Langmuir, 17(23), 7414-7424.
Picart, C., Mutterer, J., Richert, L., Luo, Y., Prestwich, G. D., Schaaf, P., et al. (2002). Molecular
basis for the explanation of the exponential growth of polyelectrolyte multilayers.
Proceedings of the National Academy of Sciences of the United States of America,
99(20), 12531-12535.
225

Planes, J., Wolter, A., Cheguettine, Y., Pron, A., Genoud, F., & Nechtschein, M. (1998).
Transport properties of polyaniline-cellulose-acetate blends. Physical Review B
(Condensed Matter), 58(12), 7774-7785.
Potschke, P., & Paul, D. R. (2003). Formation of co-continuous structures in melt-mixed
immiscible polymer blends. Journal of Macromolecular Science - Polymer Reviews,
43(1), 87-141.
Pramanik, P. K., Khastgir, D., & Saha, T. N. (1992). Conductive nitrile rubber composite
containing carbon fillers: Studies on mechanical properties and electrical conductivity.
Composites, 23(3), 183-191.
Pratt, C. (2001). Applications of Conducting Polymers
Privalko, V. P., Ponomarenko, S. M., Privalko, E. G., Lobkov, S. V., Rekhteta, N. A., Pud, A.
A., et al. (2005). Structure/property relationships for poly(vinylidene fluoride)/doped
polyaniline blends. Journal of Macromolecular Science - Physics, B44(Copyright 2006,
IEE), 749-759.
Pron, A., & Rannou, P. (2002). Processible conjugated polymers: From organic semiconductors
to organic metals and superconductors. Progress in Polymer Science (Oxford), 27(1),
135-190.
Pron, A., Zagorska, M., Nicolau, Y., Genoud, F., & Nechtschein, M. (1997). Highly conductive
composites of polyaniline with plasticized cellulose acetate, Snowbird, UT, USA.
Pud, A., Ogurtsov, N., Korzhenko, A., & Shapoval, G. (2003). Some aspects of preparation
methods and properties of polyaniline blends and composites with organic polymers.
Progress in Polymer Science (Oxford), 28(12), 1701-1753.
Qinghua, Z., Xianhong, W., Yanhou, G., Dajun, C., & Xiabin, J. (2002). Morphology and
thermal properties of conductive polyaniline/polyamide composite films. Journal of
Polymer Science, Part B (Polymer Physics), 40(22), 2531-2538.
R. Gunawidjaja, C. J. S. P. M. O. S. S. V. V. T. (2006). Flexible and Robust 2D Arrays of Silver
Nanowires Encapsulated within Freestanding Layer-by-Layer Films. Advanced
Functional Materials, 16(15), 2024-2034.
Rahmy, T. R. (2001). Action of cobra venom on the renal cortical tissues: electron microscopic
studies. Journal of Venomous Animals and Toxins, 7, 85-112.
Ram, M. K., Salerno, M., Adami, M., Faraci, P., & Nicolini, C. (1999). Physical properties of
polyaniline films: assembled by the layer-by-layer technique. Langmuir, 15(4), 1252-
1259.
Raposo, M., & Oliveira Jr, O. N. (2002). Adsorption of poly(o-methoxyaniline) in layer-by-layer
films. Langmuir, 18(18), 6866-6874.
Ravati, S. (2010). Novel Conductive Polymer Blends. PhD thesis, École Polytechnique de
Montréal, Montreal.
Ravati, S., & Favis, B. D. (2010). Polymer, Submitted.
226

Reghu, M., Yoon, C. O., Yang, C. Y., Moses, D., Smith, P., Heeger, A. J., et al. (1994).
Transport in polyaniline networks near the percolation threshold. Physical Review B,
50(Copyright (C) 2009 The American Physical Society), 13931.
Reif, J. H., Sahu, S., & Yin, P. (2006). Complexity of graph self-assembly in accretive systems
and self-destructible systems, London, Ont., Canada.
Reignier, J., & Favis, B. D. (2000). Control of the Subinclusion Microstructure in
HDPE/PS/PMMA Ternary Blends. Macromolecules, 33(19), 6998-7008.
Reignier, J., & Favis, B. D. (2003a). Core-shell structure and segregation effects in composite
droplet polymer blends. AIChE Journal, 49(4), 1014-1023.
Reignier, J., & Favis, B. D. (2003b). On the presence of a critical shell volume fraction leading
to pseudo-pure droplet behavior in composite droplet polymer blends. Polymer, 44(18),
5061-5066.
Reignier, J., Favis, B. D., & Heuzey, M.-C. (2003). Factors influencing encapsulation behavior
in composite droplet-type polymer blends. Polymer, 44(1), 49-59.
Rodrigues, P. C., & Akcelrud, L. (2003). Networks and blends of polyaniline and polyurethane:
Correlations between composition and thermal, dynamic mechanical and electrical
properties. Polymer, 44(22), 6891-6899.
Rodrigues, P. C., Lisboa-Filho, P. N., Mangrich, A. S., & Akcelrud, L. (2005).
Polyaniline/polyurethane networks. II. A spectroscopic study. Polymer, 46(7), 2285-
2296.
Roman, L. S., Andersson, M. R., Yohannes, T., & Inganás, O. (1997). Photodiode performance
and nanostructure of polythiophene/ C60 blends. Advanced Materials, 9(15), 1164-1168.
Rosner, R. B. (2000). Conductive Materials for ESD Applications: An Overview
Roy, X., Sarazin, P., & Favis, B. D. (2006). Ultraporous nanosheath materials by layer-by-layer
deposition onto co-continuous polymer-blend templates. Advanced Materials, 18(8),
1015-1019.
Rudra, J. S., Dave, K., & Haynie, D. T. (2006). Antimicrobial polypeptide multilayer
nanocoatings. Journal of Biomaterials Science, Polymer Edition, 17(11), 1301-1315.
Ruokolainen, J., Eerikainen, H., Torkkeli, M., Serimaa, R., Jussila, M., & Ikkala, O. (2000).
Comb-Shaped Supramolecules of Emeraldine Base Form of Polyaniline Due to
Coordination with Zinc Dodecyl Benzenesulfonate and Their Plasticized Self-Organized
Structures. Macromolecules, 33(25), 9272-9276.
Sarazin, P., & Favis, B. D. (2003). Morphology control in co-continuous poly(Llactide)/polystyrene
blends: A route towards highly structured and interconnected
porosity in poly(L-lactide) materials. Biomacromolecules, 4(6), 1669-1679.
Sarazin, P., Roy, X., & Favis, B. D. (2004). Controlled preparation and properties of porous
poly(L-lactide) obtained from a co-continuous blend of two biodegradable polymers.
Biomaterials, 25(28), 5965-5978.
227

Sau, K., Chaki, T., & Khastgir, D. (1997). Conductive rubber composites from different blends
of ethylene-propylene-diene rubber and nitrile rubber. Journal of Materials Science,
32(21), 5717-5724.
Sau, K. P., Chaki, T. K., & Khastgir, D. (1998). Carbon fibre filled conductive composites based
on nitrile rubber (NBR), ethylene propylene diene rubber (EPDM) and their blend.
Polymer, 39(25), 6461-6471.
Scher, H., & Zallen, R. (1970). Critical Density in Percolation Processes. The Journal of
Chemical Physics, 53(9), 3759-3761.
Schlenoff, J. B., & Dubas, S. T. (2001). Mechanism of Polyelectrolyte Multilayer Growth:
Charge Overcompensation and Distribution. Macromolecules, 34(3), 592-598.
Schlenoff, J. B., Ly, H., & Li, M. (1998). Charge and mass balance in polyelectrolyte
multilayers. Journal of the American Chemical Society, 120(30), 7626-7634.
Schmidt, V. M., Tegtmeyer, D., & Heitbaum, J. (1995). Transport of protons and water through
polyaniline membranes studied with on-line mass spectrometry. Journal of
Electroanalytical Chemistry, 385(2), 149-155.
Schmitt, J., Gruenewald, T., Decher, G., Pershan, P. S., Kjaer, K., & Loesche, M. (1993).
Internal structure of layer-by-layer adsorbed polyelectrolyte films: A neutron and X-ray
reflectivity study. Macromolecules, 26(25), 7058-7063.
Schneider, G., & Decher, G. (2004). From Functional Core/Shell Nanoparticles Prepared via
Layer-by-Layer Deposition to Empty Nanospheres. Nano Letters, 4(10), 1833-1839.
Schoeler, B., Kumaraswamy, G., & Caruso, F. (2002). Investigation of the influence of
polyelectrolyte charge density on the growth of multilayer thin films prepared by the
layer-by-layer technique. Macromolecules, 35(3), 889-897.
Schwarz, S., Eichhorn, K. J., Wischerhoff, E., & Laschewsky, A. (1999). Polyelectrolyte
adsorption onto planar surfaces: A study by streaming potential and ellipsometry
measurements. Colloids and Surfaces A: Physicochemical and Engineering Aspects,
159(2-3), 491-501.
Scott, R. L. (1949). The Thermodynamics of High Polymer Solutions. IV. Phase Equilibria in the
Ternary System: Polymer---Liquid 1---Liquid 2. The Journal of Chemical Physics, 17(3),
268-279.
Segal, E., Haba, Y., Narkis, M., & Siegmann, A. (2001). On the structure and electrical
conductivity of polyaniline/polystyrene blends prepared by an aqueous-dispersion
blending method. Journal of Polymer Science, Part B: Polymer Physics, 39(5), 611-621.
Shacklette, L. W., Han, C. C., & Luly, M. H. (1993). Polyaniline blends in thermoplastics.
Synthetic Metals, 57(1), 3532-3537.
Shimazaki, Y., Mitsuishi, M., Ito, S., & Yamamoto, M. (1997). Preparation of the Layer-by-
Layer Deposited Ultrathin Film Based on the Charge-Transfer Interaction. Langmuir,
13(6), 1385-1387.
228

Shirakawa, H., Louis, E. L., MacDiarmid, A. G., Chiang, C. K., & Heeger, A. J. (1977).
Synthesis of electrically conducting organic polymers: halogen derivatives of
polyacetylene, (CH)x. J. Chem. Soc., Chem. Commun., 578 - 580.
Shiratori, S. S., & Rubner, M. F. (2000). pH-Dependent Thickness Behavior of Sequentially
Adsorbed Layers of Weak Polyelectrolytes. Macromolecules, 33(11), 4213-4219.
Sichel, E. (1982). Carbon Black-Polymer Composites. New York: Marcel Dekker.
Siggia, E. D. (1979). Late stages of spinodal decomposition in binary mixtures. Physical Review
A (General Physics), 20(2), 595-605.
Skorokhov, V. V. (1972). The Rheological Principles of Sintering Theory. Kiev: Nauka.
Smuckler J, H., & Finnerty P, M. (1974). Performance of Conductive Carbon Blacks in a Typical
Plastics System Fillers and Reinforcements for Plastics (pp. 171-183). Washington, D.
C.: American Chemical Society.
Soares, B. G., Gubbels, F., Jerome, R., Teyssie, P., Vanlathem, E., & Deltour, R. (1995).
Electrical conductivity in carbon black-loaded polystyrene-polyisoprene blends. Selective
localization of carbon black at the interface. Polymer Bulletin (Berlin), 35(1-2), 223.
Solis, F. J., Cruz, d. l., & M., O. (1999). Surface-induced layer formation in polyelectrolytes.
Journal of Chemical Physics, 110(23), 11517-11522.
Solonin, S. M., & Chernyshev, L. I. (1975). Effects of sintering conditions on the
physicomechanical properties of porous tungsten. Soviet Powder Metallurgy and Metal
Ceramics (English translation of Poroshkovaya Metallurgiya), 14(10), 806-808.
Stauffer, D., & Aharony, A. (1994b). Introduction to Percolation Theory. London: Taylor and
Francis.
Steinmann, S., Gronski, W., & Friedrich, C. (2001). Cocontinuous polymer blends: Influence of
viscosity and elasticity ratios of the constituent polymers on phase inversion. Polymer,
42(15), 6619-6629.
Steinmann, S., Gronski, W., & Friedrich, C. (2002). Quantitative rheological evaluation of phase
inversion in two-phase polymer blends with cocontinuous morphology. Rheologica Acta,
41(1-2), 77-86.
Stockton, W. B., & Rubner, M. F. (1997). Molecular-level processing of conjugated polymers. 4.
Layer-by-layer manipulation of polyaniline via hydrogen-bonding interactions.
Macromolecules, 30(9), 2717-2725.
Stratouras, G., & Kosmas, M. (1992). Are ring polymers adsorbed on a surface more than linear
polymers? Macromolecules, 25(12), 3307-3308.
Su, T. M., Ball, I. J., Conklin, J. A., Huang, S.-C., Larson, R. K., Nguyen, S. L., et al. (1997).
Polyaniline/polyimide blends for pervaporation and gas separation studies. Synthetic
Metals, 84(1-3 pt 1), 801-802.
Sui, Z., Salloum, D., & Schlenoff, J. B. (2003). Effect of Molecular Weight on the Construction
of Polyelectrolyte Multilayers:&nbsp; Stripping versus Sticking. Langmuir, 19(6), 2491-
2495.
229

Sumita, M., Kayaki, H., & Miyasaka, K. (1986). Effect of Melt Viscosity and Surface Tension of
Polymers on the Percolation Threshold of Conductive-Particle-Filled Polymeric
Composites. Journal of Macromolecular Science - Physics, B25(1-2), 171-184.
Sumita, M., Sakata, K., Asai, S., Miyasaka, K., & Nakagawa, H. (1991). Dispersion of fillers and
the electrical conductivity of polymer blends filled with carbon black. Polymer Bulletin
(Berlin), 25(2), 265-271.
Sumita, M., Sakata, K., Hayakawa, Y., Asai, S., Miyasaka, K., & Tanemura, M. (1992). Double
percolation effect on the electrical conductivity of conductive particles filled polymer
blends. Colloid & Polymer Science, 270(2), 134-139.
Sundararaj, U., & Macosko, C. W. (1995). Drop breakup and coalescence in polymer blends: the
effects of concentration and compatibilization. Macromolecules, 28(Compendex), 2647-
2657.
Tang, Z., Kotov, N. A., & Giersig, M. (2002). Spontaneous Organization of Single CdTe
Nanoparticles into Luminescent Nanowires. Science, 297(5579), 237-240.
Tchomakov, K. P., Favis, B. D., Huneault, M. A., Champagne, M. F., & Tofan, F. (2004).
Composite droplets with core/shell morphologies prepared from HDPE/PS/PMMA
ternary blends by twin-screw extrusion. Polymer Engineering and Science, 44(4), 749-
759.
Tchoudakov, R., Breuer, O., Narkis, M., & Siegmann, A. (1996). Conductive polymer blends
with low carbon black loading: Polypropylene/polyamide. Polymer Engineering &
Science, 36(10), 1336-1346.
Tessler, N., Denton, G. J., & Friend, R. H. (1996). Lasing from conjugated-polymer
microcavities. Nature, 382(6593), 695-697.
Tjahjadi, M., Ottino, J. M., & Stone, H. A. (1994). Estimating interfacial tension via relaxation
of drop shapes and filament breakup. AIChE Journal, 40(3), 385-394.
Tjipto, E., Quinn, J. F., & Caruso, F. (2007). Layer-by-layer assembly of weak-strong copolymer
polyelectrolytes: a route to morphological control of thin films. Journal of Polymer
Science, Part A (Polymer Chemistry), 45(Copyright 2008, The Institution of Engineering
and Technology), 4341-4351.
Tomotika, S. (1935a). Instability of a cylindrical thread of a viscous liquid surrounded by
another viscous fluid. Proceedings of the Royal Society of London, 150A, 322-337.
Tomotika, S. (1935b). On the instability of a cylindrical thread of a viscous liquid surrounded by
another viscous liquid. Proc. Roy. Soc., A150, 322-337.
Torza, S., & Mason, S. G. (1970). Three-phase interactions in shear and electrical fields. Journal
of Colloid and Interface Science, 33(1), 67-83.
Tse, C. W., Man, K. Y. K., Cheng, K. W., Mak, C. S. K., Chan, W. K., Yip, C. T., et al. (2007).
Layer-by-layer deposition of rhenium-containing hyperbranched polymers and
fabrication of photovoltaic cells. Chemistry - A European Journal, 13(1), 328-335.
Urashita, S., Kawakatsu, T., & Doi, M. (2000). Morphologies of multiphase polymer blend
systems, Kanazawa, Japan.
230

Utracki, L. A. (1991). On the viscosity-concentration dependence of immiscible polymer blends.
Journal of Rheology, 35(8), 1615-1637.
Utracki, L. A. (1998). Commercial Polymer Blends Springer - Verlag.
Valera, T. S., Morita, A. T., & Demarquette, N. R. (2006). Study of morphologies of
PMMA/PP/PS ternary blends. Macromolecules, 39(7), 2663-2675.
Van Lent, B., Scheutjens, J., & Cosgrove, T. (1987). Self-consistent field theory for the
adsorption of ring polymers from solution. Macromolecules, 20(2), 366-370.
Van Oene, H. J. (1972). Modes of dispersion of viscoelastic fluids in flow. Colloid Interface Sci.,
40, 448-467.
Vanoene, H. (1972). Modes of dispersion of viscoelastic fluids in flow. Journal of Colloid and
Interface Science, 40(3), 448-467.
Veenstra, H., Norder, B., van Dam, J., & Posthuma de Boer, A. (1999). Stability of cocontinuous
polystyrene/poly(ether-ester) blends in shear flow. Polymer, 40(18), 5223-
5226.
Veenstra, H., Van Dam, J., & Posthuma de Boer, A. (2000). On the coarsening of co-continuous
morphologies in polymer blends: Effect of interfacial tension, viscosity and physical
cross-links. Polymer, 41(8), 3037-3045.
Veenstra, H., Verkooijen, P. C. J., van Lent, B. J. J., van Dam, J., de Boer, A. P., & Nijhof, A. P.
H. J. (2000). On the mechanical properties of co-continuous polymer blends:
Experimental and modelling. Polymer, 41(5), 1817-1826.
Verhelst, W. F., Wolthuis, K. G., Voet, A., Ehrburger, P., & Donnet, J. B. (1977). The role of
morphology and structure of carbon blacks in the electrical conductance of vulcanizates.
Rubber Chem. Technol., 50, 735-746.
Vikki, T., Pietila, L.-O., Osterholm, H., Ahjopalo, L., Takala, A., Toivo, A., et al. (1996).
Molecular Recognition Solvents for Electrically Conductive Polyaniline.
Macromolecules, 29(8), 2945-2953.
Virgilio, N., Desjardins, P., L'Esperance, G., & Favis, B. D. (2009). In situ measure of interfacial
tensions in ternary and quaternary immiscible polymer blends demonstrating partial
wetting. Macromolecules, 42(19), 7518-7529.
Virgilio, N., Favis, B. D., Pepin, M.-F., Desjardins, P., & L'Esperance, G. (2005a). High contrast
imaging of interphases in ternary polymer blends using focused ion beam preparation and
atomic force microscopy. Macromolecules, 38(Compendex), 2368-2375.
Virgilio, N., Marc-Aurele, C., & Favis, B. D. (2009). Novel self-assembling close-packed droplet
array at the interface in ternary polymer blends. Macromolecules, 42(Compendex), 3405-
3416.
Volodkin, D. V., Petrov, A. I., Prevot, M., & Sukhorukov, G. B. (2004). Matrix Polyelectrolyte
Microcapsules: New System for Macromolecule Encapsulation. Langmuir, 20(8), 3398-
3406.
Wallin, T., & Linse, P. (1996). Monte Carlo simulations of polyelectrolytes at charged micelles.
1. Effects of chain flexibility. Langmuir, 12(2), 305.
231

Wang Gusukuma, L., Prates, J. C., & Smith, R. L. (2004). Chorade Tendineae Architecture in the
Papillary Muscle Insertion. International Journal of Morphology, 22, 267-272.
Wang, Y., & Rubner, M. F. (2002). Electrically conductive semiinterpenetrating polymer
networks of poly(3-octylthiophene). Macromolecules, 25(12), 3284-3290.
Watcharaphalakorn, S., Ruangchuay, L., Chotpattananont, D., Sirivat, A., & Schwank, J. (2005).
Polyaniline/polyimide blends as gas sensors and electrical conductivity response to CO-
N2 mixtures. Polymer International, 54(8), 1126-1133.
Wessling, B. (1994). Passivation of metals by coating with polyaniline: Corrosion potential shift
and morphological changes. Advanced Materials, 6(3), 226-228.
Willemse, R. C., de Boer, A. P., van Dam, J., & Gotsis, A. D. (1998). Co-continuous
morphologies in polymer blends: A new model. Polymer, 39(24), 5879-5887.
Willemse, R. C., Posthuma de Boer, A., van Dam, J., & Gotsis, A. D. (1999). Co-continuous
morphologies in polymer blends: the influence of the interfacial tension. Polymer, 40(4),
827-834(828).
Willemse, R. C., Ramaker, E. J. J., van Dam, J., & Posthuma de Boer, A. (1999). Morphology
development in immiscible polymer blends: initial blend morphology and phase
dimensions. Polymer, 40(24), 6651-6659.
Wise, D. L., Wnek, G. E., Trantolo, D. J., Cooper, T. M., & Gresser, J. D. (1997). Electrical and
optical polymer systems. New York: Marcel Dekker.
Wolter, A., Banka, E., Genoud, F., Pron, A., & Nechtschein, M. (1997). Electronic transport in
composites based on polyaniline blended with cellulose acetate - a DC-conductivity and
ESR analysis. Synthetic Metals, 84(1-3 pt 1), 753-754.
Won-Jung, K., Tae-Young, K., Jung-Woo, K., Yun-Sang, K., Chang-Mo, P., & Suh, K. S.
(2004). Charge transport phenomena of polyaniline-DBSA/polystyrene blends.
Transactions of the Korean Institute of Electrical Engineers, C, 53(6), 305-311.
Wong, C. P. (2003-2008). Fundamental Understanding and Performance Enhancement of
Conductive Adhesive for Microelectronic Packaging Applications. Georgia Institute of
Technology
Woo Jin, B., Won Ho, J., & Yun Heum, P. (2003). Preparation of polystyrene/polyaniline blends
by in situ polymerization technique and their morphology and electrical property.
Synthetic Metals, 132(3), 239-244.
Wu, S. (1971). J. Polym. Sci., C34, 19.
Wu, S. (1982). Polymer Interface and Adhesion. New York: Marcel Dekker Inc.
Yang, C. Y., Cao, Y., Smith, P., & Heeger, A. J. (1993a). Morphology of conductive, solutionprocessed
blends of polyaniline and poly(methyl methacrylate). Synthetic Metals, 53(3),
293-301.
Yang, C. Y., Cao, Y., Smith, P., & Heeger, A. J. (1993b). Morphology of polyaniline in solution
processed blends with poly(methyl methacrylate). Polymer Preprints, Division of
Polymer Chemistry, American Chemical Society, 34(1), 790.
232

Yasuda, K., Armstrong, R., & Cohen, R. (1981). Shear flow properties of concentrated solutions
of linear and star branched polystyrenes. Rheologica Acta, 20(2), 163-178.
Yipeng Sun, X. Z. C. S. B. W. J. S. (1996). Fabrication of ultrathin film containing bienzyme of
glucose oxidase and glucoamylase based on electrostatic interaction and its potential
application as a maltose sensor. Macromolecular Chemistry and Physics, 197(1), 147-
153.
Yoshikawa, H., Hino, T., & Kuramoto, N. (2006). Effect of temperature and moisture on
electrical conductivity in polyaniline/polyurethane (PANI/PU) blends. Synthetic Metals,
156(18-20), 1187-1193.
Yuan, Z., & Favis, B. D. (2004). Macroporous poly(L-lactide) of controlled pore size derived
from the annealing of co-continuous polystyrene/poly(L-lactide) blends. Biomaterials,
25(11), 2161-2170.
Yuan, Z., & Favis, B. D. (2006). Influence of the efficacy of interfacial modification on the
coarsening of cocontinuous PS/HDPE blends during quiescent annealing. Journal of
Polymer Science, Part B: Polymer Physics, 44(4), 711-721.
Zasadzinski, J. A., Viswanathan, R., Madsen, L., Garnaes, J., & Schwartz, D. K. (1994).
Langmuir-Blodgett films. Science, 263(5154), 1726-1733.
Zhang, J., Ravati, S., Virgilio, N., & Favis, B. D. (2007). Ultralow percolation thresholds in
ternary cocontinuous polymer blends. Macromolecules, 40(25), 8817-8820.
Zhang, Q., Jin, H., Wang, X., & Jing, X. (2001). Morphology of conductive blend fibers of
polyaniline and polyamide-11. Synthetic Metals, 123(3), 481-485.
Zhenhua Yuan, B. D. F. (2005). Coarsening of immiscible co-continuous blends during
quiescent annealing. AIChE Journal, 51(1), 271-280.
Zilberman, M., Siegmann, A., & Narkis, M. (1998). Polyaniline containing ternary immiscible
polymer blends: A melt processable conductive system, Anaheim, CA, USA.
Zilberman, M., Siegmann, A., & Narkis, M. (2000a). Electrically conductive extruded filaments
of polyaniline/polymer blends. Journal of Polymer Engineering, 20(2), 97-110.
Zilberman, M., Siegmann, A., & Narkis, M. (2000c). Melt-processed, electrically conductive
ternary polymer blends containing polyaniline. Journal of Macromolecular Science -
Physics, B39(3), 333-347.
Zilberman, M., Titelman, G. I., Siegmann, A., Haba, Y., Narkis, M., & Alperstein, D. (1997).
Conductive blends of thermally dodecylbenzene sulfonic acid-doped polyaniline with
thermoplastic polymers. Journal of Applied Polymer Science, 66(2), 243-253.
233

APPENDIX I
A- 1 Ultra-Low Percolation Thresholds in Ternary Co-
Continuous Polymer Blends
Jianhong Zhang, Sepehr Ravati, Nick Virgilio, and Basil D. Favis ∗
CREPEC, Department of Chemical Engineering, École Polytechnique de Montréal, Montréal,
Québec, H3C 3A7 Canada
Published in Macromolecules in December 2007
A-1.1. Introduction. In binary immiscible polymer blends, as the concentration of the minor
component increases, the structure percolates and its continuity also increases (Potschke & Paul,
2003; Reignier & Favis, 2000). At higher concentrations, closely associated with the region of
phase inversion, a co-continuous morphology is obtained and is characterized by two fully
continuous phases. By selectively controlling the interface, composition, processing temperature,
shear rate and annealing time of the blends, it is possible to control the pore size of the cocontinuous
network over 2-3 orders of magnitude (Yuan & Favis, 2004, 2006). Co-continuous
polymer blends have the potential of opening particular application fields where the presence of
interconnected structures are a necessary feature (as in separation phenomena, electrical
conductivity, tissue engineering scaffolds and drug delivery devices). A significant body of
∗ Corresponding author. E-mail: basil.favis@polymtl.ca Tel: +1-(514)3404711 ext.4527. Fax:
+1-(514)3404159
234

literature has examined factors influencing the percolation threshold in binary co-continuous
polymer blends and also the width and concentration range of the region of dual-phase continuity
(Yuan & Favis, 2004, 2006). Most studies on binary systems report percolation thresholds in a
range of about 15-20% concentration of minor component. This significant amount of the second
phase, necessary to obtain a co-continuous structure, considerably limits the range of possible
applications. However, very few studies have examined the continuity phenomena in multiphase
(more than two phases) systems. The possibility of dramatically decreasing the percolation
threshold of one component in order to achieve co-continuity at concentration levels of a few
percent would be an important achievement. A few studies have contributed towards attaining
such a result. Jérôme and co-workers reported on the preparation of a co-continuous polymer
blend with the addition of a solid filler showing a very low percolation threshold for the filler at
about 0.2 volume percent. Interestingly, this filler was located at the interface in a chain-like
structure shape (Gubbels, et al., 1994; Soares, et al., 1995). However, the limitation in that case
is that the particle is not thermodynamically stable at the interface. It has a particular affinity for
one of the phases and is simply transiting from one phase to the other. Jorgensen and Utracki
report very low percolation thresholds for polyethylene in polystyrene binary blends of very high
molecular weight (Lyngaae-Jorgensen & Utracki, 2003). In other studies, Narkis, Zilberman and
coworkers (Narkis, et al., 2000a; Zilberman, Siegmann, & Narkis, 2000c) achieved multiple
percolation by dispersing a third phase into one of the phases in a co-continuous binary blend.
This ultimately creates a third percolated system, but the percolation thresholds are relatively
high at 8-12% composition of this third phase.
A potentially interesting way to achieve low percolation threshold materials is to combine a cocontinuous
structure with a composite droplet type of morphology. The composite droplet
morphology is a matrix/dispersed phase system in which the dispersed phase has a droplet-in
droplet structure. Although little work has been published on polymeric versions of such
systems, a number of key papers have appeared (Berger, Kammer, & Kummerlowe, 1984;
Omonov, Harrats, & Groeninckx, 2005; Reignier & Favis, 2003a; Reignier, et al., 2003; Van
Oene, 1972). The spreading coefficients (Torza & G. Mason, 1970) (Equation A-1.1) and
variations on that concept have been successfully used (Hobbs, et al., 1988; Reignier & Favis,
2000; Valera, et al., 2006) to predict the thermodynamically stable preferential encapsulation of
one polymeric phase by another. Encapsulation is governed by the various interfacial tension
235

pairs within a ternary polymer blend and is related to the minimization of the interfacial free
energy. In Equation A-1.1,
Equation A-1.1
λ = γ − γ − γ
ij
jk
λij is defined as the spreading coefficient describing the tendency of component i to encapsulate
or spread around component j in a matrix of component k. γjk, γik and γij are the interfacial
tensions of the different polymer pairs. In blends of one major constituent (polymer A) and two
minor ones (polymers B and C, the dispersed phases), four different morphologies are possible
(Figure A-1.1) : a) separate dispersions; b) partial encapsulation and c), d) two possibilities of
complete wetting (composite droplets). Each morphology has a distinct set of spreading
coefficients.
Figure A-1.1. Morphologies of ternary polymer blends containing one major phase (polymer A
(white)) and two minor ones (polymers B (gray) and C (black)): (a) separate dispersions; (b)
partial encapsulation; (c and d) composite droplets (complete wetting)
Recently, in our group, studies have been conducted on ternary immiscible polymer blends of
high-density polyethylene (HDPE), polystyrene (PS) and poly(methyl methacrylate) (PMMA)
ik
ij
236

(Reignier & Favis, 2000, 2003a; Reignier, et al., 2003). When PS and PMMA are dispersed in a
HDPE matrix, composite droplets constituted of a well-segregated PMMA core and PS shell are
formed. By controlling the relative amounts of PS and PMMA, it is possible to control the
thickness of the PS shell down to 30-40 nm, near the range of a molecular layer of PS.
Furthermore, virtually all of the PMMA is present as sub-inclusions within the PS dispersed
phase after approximately 2 minutes of mixing.
This paper reports on the development of ternary percolated co-continuous systems in
HDPE/PS/PMMA blends in which HDPE and PMMA form two continuous networks, while the
PS forms a continuous sheath structure at the HPDE/PMMA interface (Figure A-1.2). By
controlling the relative amounts of HDPE, PS and PMMA, it will be shown that it is possible to
decrease the percolation threshold of the PS layer down to less than 3%, resulting in a PS
continuous network of ultra low composition.
Figure A-1.2. Schematic of the HDPE/PS/PMMA multiple percolated structure. a) Threedimensional
view of the blend structure, showing the thin PS layer at the HDPE/PMMA
interface. b) Section showing the relative position of the different phases.
237

A-1.2. Experimental Section
A-1.2.1. Materials: HDPE 04352N and PS 615APR were both obtained from the Dow Chemical
Company. PMMA 200336 was purchased from Sigma-Aldrich. An antioxidant, Irganox B225
from Ciba-Geigy, was added to the mixture to reduce thermal oxidation of polyethylene (0.2
weight percent). Physical properties of the materials were measured previously (Reignier &
Favis, 2003a) and are given in Table A-1.1.
Table A-1.1. Homopolymer properties
Polymers M w a ×10-
3 g/mol
M n a ×10-3
g/mol
Melt Index b
ASTM
g/10min
Density b g/cm3 at
20°C 200°C
η 1* × 10-3 c Pa.s
at 200°C
N 1 × 10-4 d Pa at
200°C
HDPE 79 24 4 0.96 0.75 1.2 1.2 2.2 2.2
PS 290 141 15 1.04 0.97 1.7 2.6 11 5.1
PMMA 11.9 7.8 - - 1 0.04 0.04 0.02 0.6
a Measured by GPC, b Obtained from supplier, c η * is the complex viscosity, d N1 is the first normal stress difference, e At average shear rate
during blending:
1
25 −
•
γ = s , f 4
At average shear stress during blending: τ = 2 . 7×
10 Pa
A-1.2.1. Measurement of the interfacial tensions and calculation of the spreading coefficients:
The interfacial tensions were measured using the breaking thread method, as described in an
earlier publication (Reignier & Favis, 2000). The different interfacial tensions and spreading
coefficients for the HDPE/PS/PMMA ternary blends are given in Table A-1.2.
γ&
e
f
τ γ& τ
238

Table A-1.2. Interfacial tensions and spreading coefficients for HDPE/PS/PMMA blends at
200°C.
Polymer pairs
%
Interfacial tension
Continuity
γ (mN/m)
m
=
m
initial
− m
( i )
initial final ×
Polymer pairs
Spreading of i on j (i/j)
100
Spreading coefficients
λ (mN/m)
HDPE/PS 5.1 PS/PMMA 1.1
PS/PMMA 2.4 PMMA/PS -5.9
HDPE/PMMA 8.6 HDPE/PS -11.3
A-1.2.3. Blend preparation: Prior to melt mixing, all polymers were pre-dried for 24h at 80°C in
a vacuum oven to remove any moisture from the pellets or powders. Ternary blends were
prepared by melt-mixing in a 50 ml Haake internal mixer with twin cam rotors, with the
temperature of the mixing chamber initially set at 200°C and rotor speed set at 50 rpm. After
mixing for 7 minutes under a constant flow of dry nitrogen, samples were immediately quenched
in a bath of liquid nitrogen to freeze-in the morphology.
A-1.2.4. Continuity measurement: Solvent extraction has been used to determine the
composition region of co-continuity of the blends. As a primary advantage, solvent extraction is
an absolute measurement (Steinmann, et al., 2001) and is capable of detecting the existence of
co-continuous microstructures when the components are soluble in specific solvents (Potschke &
Paul, 2003). The level of continuity of a phase in a sample is given by Equation A-1.2:
Equation A-1.2
In this equation, minitial is the initial mass of the sample, mfinal is the final mass of the sample and
minitial(i) is the mass of polymer i contained in the sample before selective extraction, calculated
by considering the blend as homogeneous. Selective solvent extraction of the PMMA and PS
phases were performed respectively in acetic acid and cyclohexane in a soxhlet extraction
apparatus. Mass loss measurements were carried out for one week and were used to calculate the
extent of continuity of the PMMA and PS phases using Equation A-1.2.
A-1.2.5. Focused Ion Beam (FIB) preparation and Atomic Force Microscopy (AFM)
morphology analysis: The specimens were initially cryomicrotomed to create a plane face using
239

a Leica RM2165 microtome equipped with a cooling chamber. FIB surface etching was
performed using a Hitachi 2000A Ga + focused ion beam operated at 30 keV and 0.8 nA, with a
dwelling time of 3 μsec. The etched surface was then analyzed by tapping mode atomic force
microscopy (AFM) using a scanning probe microscope Dimension 3100 from Veeco Instruments
equipped with a Nanoscope IVa controller. Silicon tips, model PPP-NCH-W from Nanosensors,
with a force constant of 10-130 N/m (nominal value of 42 N/m) and resonance frequency of 204-
497 kHz (nominal value of 330 kHz), were used. The cantilever was oscillated at approximately
98% of the resonance frequency and topographic (height) images were taken at approximately
95% of the free oscillating amplitude. The samples were fixed on a metallic support using
graphite tape. This procedure is thoroughly described in a previous paper (Virgilio, Favis, Pepin,
Desjardins, & L'Esperance, 2005b).
A-1.3. Results and Discussion
Initially, the continuity development of the PMMA phase in binary HDPE/PMMA blends was
examined in order to determine the boundaries required to achieve co-continuity in
HDPE/PMMA. After preparation by melt-mixing, the blends were immediately quenched in
liquid nitrogen to freeze-in the morphology. To measure the continuity of the PMMA phase in
each blend, a selective solvent extraction of the PMMA phase, using acetic acid, was performed.
As the PMMA volume fraction increases from 30% to 40%, the continuity increases very rapidly
from 17% to 95%. A further increase of the PMMA volume fraction to 45% yields a 98%
continuous PMMA phase and results in a binary co-continuous polymer blend.
Subsequently, ternary HDPE/PS/PMMA blends were prepared at different compositions
(given in Table A-1.3) based on the co-continuity region found for HDPE/PMMA binary blends.
The qualitative microstructural features of the blends were analyzed using a Focused Ion Beam
(FIB)-Atomic Force Microscopy (AFM) method described in a previous article (Virgilio, et al.,
2005b). FIB etching involves the interaction of a gallium ion beam with the blend surface and is
a surface preparation step prior to the AFM surface analysis. Since the gallium ion beam has
significantly different etching rates for each of the individual components, this etching process
induces a topological contrast between the different phases, each surface level corresponding to a
specific phase. Observation of these surfaces by AFM in the topological mode allows for the
240

acquisition of images with a very high contrast. Since all the three polymers are immiscible, the
three different levels observed on the AFM images each correspond to a given polymer. In
HDPE/PS/PMMA ternary blends, PS is the less etched material, and thus corresponds to the
high-level domains, while HDPE corresponds to the mid-level domains and PMMA to the lowlevel
ones. This method of analysis is particularly useful for multicomponent blends and for fine
structure analysis since the etching process is sensitive to the nature of the materials and
eliminates some artifacts related to classical microtomy and staining preparation methods.
a) b)
HDPE
PMMA
PS layer
PS droplet
PMMA
PS layer
PS droplet
0 nm 65 nm 0 nm 65 nm
HDPE
5 5
Figure A-1.3. FIB-AFM images of HDPE/PS/PMMA ternary blends: a) 50/10/40(% vol.) and b)
50/12.5/37.5. Images were taken at 95% of the free oscillation amplitude. Scan sizes are 20 μm x
20 μm. The white line in Figure A-1.3b indicates the section analyzed below.
241

Figure A-1.3 shows AFM images after FIB treatment of the microstructure of ternary blends
No.5 and No.6, as described in Table A-1.3. The HDPE (mid-level domains, notice the fine
texture corresponding to the crystalline structure) and PMMA phases (low-level domains, in dark
orange) appear as the main components and are separated by a thin PS layer (high-level domains,
in yellow). PS/PMMA and PS/HDPE composite droplets can also be seen in the HDPE and
PMMA phases respectively, while some pure dispersed PS is observed in the PMMA. Finally,
small quantities of PMMA and HDPE are visible in the PS layer. Even though the segregation is
not perfect, it is clear that most of the PS is located at the interface between the PMMA and the
HDPE phases. The white bar in Figure A-1.3b) corresponds to the section analyzed below the
AFM image, quantitatively showing the relative topographical levels of each phase.
The presence of a PS layer at the interface of the co-continuous HDPE/PMMA blends
significantly reduces the PS volume fraction required for its percolation and continuity
development as compared to classical binary HDPE/PS and PMMA/PS blends. Continuity data
based on gravimetric solvent extraction clearly demonstrate this effect and are shown in Table A-
1.3 and Figure A-1.4. Cyclohexane and acetic acid were used to extract PS and PMMA
respectively. It can be seen that in this triple percolated system, a PS volume composition as low
as 3% results in a PS phase continuity of about 70%, a very high level of continuity for such a
small volume fraction of PS. Moreover, the PS layer continuity increases with increasing PS
volume content, reaching an apparent maximum value of approximately 80%. Interestingly, it
seems impossible to further increase the PS continuity by increasing its relative composition.
This upper limit in the apparent continuity of the PS is most probably due to the PS trapped as
droplets and composite droplets in the PMMA and HDPE phases as shown in Figure A-1.3.
Nevertheless, most of the PS has clearly and spontaneously structured itself at the HDPE/PMMA
interface.
242

Table A-1.3. Continuity of the PMMA and PS phases as a function of composition using the
selective solvent dissolution technique
Blend Volume fractions (%) Extracting PMMA Extracting PS
No. HDPE PMMA PS PMMA continuity (%) PS continuity (%)
1 50 50 0 103 0
2 50 47 3 103 69
3 50 45 5 102 73
4 50 42.5 7.5 103 74
5 50 40 10 102 76
6 50 37.5 12.5 104 77
Continuity (%)
100
90
80
70
60
50
40
30
20
10
0
PS composition in HDPE/PMMA/PS
PS composition in HDPE/PS
PMMA composition in HDPE/PMMA
0 5 10 15 20 25 30 35 40 45
Volume Fraction (%)
Figure A-1.4. Continuity of PS or PMMA as a function of the composition using the solvent
dissolution technique
243

A-1.4. Conclusions
Through a combination of composite droplet and co-continuous blend morphology preparation
methodologies, a continuous shell structure of polystyrene has been situated at the interface of a
co-continuous polyethylene/poly(methyl methacrylate) blend. This multiple percolated,
interfacial tension driven structure results in a dramatic decrease in the percolation threshold
volume fraction of the encapsulating PS component. As little as 3% polystyrene is shown to
generate a continuous PS network. The encapsulation effect follows the prediction based on the
spreading coefficient theory. This approach could be used as a route to generate novel cocontinuous
structures and also as a technique to significantly reduce percolation thresholds in
multiphase blends.
A-1.5. References
(1) Favis, B. D., In Polymer Blends: Formulation and Performance, Paul, D. R.; Bucknall, C.
B., Eds. John Wiley & Sons, Inc: New York, 2000; Vol. 1, pp 239-289.
(2). Potschke, P.; Paul, D. R., Journal of Macromolecular Science - Polymer Reviews 2003, 43,
(1), 87-141.
(3) Yuan, Z.; Favis, B. D., Journal of Polymer Science, Part B: Polymer Physics 2006, 44, (4),
711-721.
(4) Yuan, Z.; Favis, B. D., Biomaterials 2004, 25, (11), 2161-2170.
(5) Soares, B. G.; Gubbels, F.; Jerome, R.; Teyssie, P.; Vanlathem, E.; Deltour, R., Polymer
Bulletin (Berlin) 1995, 35, (1-2), 223.
(6) Gubbels, F.; Jerome, R.; Teyssie, P.; Vanlathem, E.; Deltour, R.; Calderone, A.; Parente, V.;
Bredas, J. L., Macromolecules 1994, 27, (7), 1972-1974.
(7) Lyngaae-Jorgensen, J.; Utracki, L. A., Polymer 2003, 44, (5), 1661-9.
(8) Zilberman, M.; Siegmann, A.; Narkis, M., Journal of Macromolecular Science - Physics
2000, B39, (3), 333-47.
(9) Narkis, M.; Haba, Y.; Segal, E.; Zilberman, M.; Titelman, G. I.; Siegmann, A., Polymers for
Advanced Technologies 2000, 11, (8-12), 665-673.
(10) Omonov, T. S.; Harrats, C.; Groeninckx, G., Polymer 2005, 46, (26), 12322-12336.
(11) Reignier, J.; Favis, B. D., AIChE Journal 2003, 49, (4), 1014-1023.
(12) Reignier, J.; Favis, B. D.; Heuzey, M.-C., Polymer 2003, 44, (1), 49-59.
(13) Berger, W.; Kammer, H. W.; Kummerlowe, C., Makromol Chem Suppl B 1984, 8, 101.
(14) Van Oene, H. J., Colloid Interface Sci. 1972, 40, 448-467.
(15) Torza, S.; Mason, G., J. Coll. Interface Sci. 1970, 33, (1), 67-83.
(16) Hobbs, S. Y.; Dekkers, M. E. J.; Watkins, V. H., Polymer 1988, 29, (9), 1598-1602.
244

(17) Reignier, J.; Favis, B. D., Macromolecules 2000, 33, (19), 6998-7008.
(18) Valera, T. S.; Morita, A. T.; Demarquette, N. R., Macromolecules 2006, 39, (7), 2663-
2675.
(19) Steinmann, S.; Gronski, W.; Friedrich, C., Polymer 2001, 42, (15), 6619-6629.
(20) Virgilio, N.; Favis, B. D.; Pepin, M.-F.; Desjardins, P.; L'Esperance, G., Macromolecules
2005, 38, (6), 2368-2375.
245

APPENDIX II
A- 2 Estimation of Conductivity of Porous Conductive
Devices by Semi-Empirical Models
Assembly of dimensional structures through the templating of multilayers onto 3D nanometer to
micrometer scale objects was one of the most outstanding strategy modifications for LbL
preparation. A few works have been devoted to developing 3D structures as the substrate in other
areas of science such as drug delivery, but to date 3D templates have not been used for
fabrication of LbL conductive multilayers. The pioneering works in which the LbL films were
assembled on a 3D colloidal core(Caruso, Caruso, et al., 1998; Caruso, Caruso, & Mohwald,
1999) and 3D nano-sized porous templates(Hou, Harrell, Trofin, Kohli, & Martin, 2004) were
carried out for controlled release, encapsulation of guest substances, and other purposes(Caruso
& Schuler, 2000; Hou, et al., 2004; Park, Deng, & Advincula, 2005; Schneider & Decher, 2004;
Volodkin, Petrov, Prevot, & Sukhorukov, 2004). Fabrication of a highly-controlled porous 3D
substrate on which the LbL multilayer can be deposited is of great importance, whereas
controlling the surface area can influence on the absorbed multilayer(Caruso, 2003). Applying a
3D structure enables us to not only manipulate the function of singular LbL multilayer films, but
to create highly-controlled complex structures with a hierarchy of order and function that can be
used to create devices and microsystems with special properties. The effect of porosity on
mechanical, physical, and electrical properties of porous materials has been studied. For sintered
metals, the presence of pores causes a decrease in strength and ductility and a detrimental effect
on conductivity and magnetic properties. Hence, the study of the relationship between
microstructure, and particularly porosity, of conductive porous samples and conductivity is of
great importance.
In this part of the work, a semi-empirical model is obtained to calculate the conductivity of
porous samples represented in Chapter 5 as a function of porosity and conductivity of the pure
246

PANI. Theoretically, the extent of the PANI phase in sample B is calculated from the mass
increase of sample B from gravimetric data represented in Figure A-2.1.
Mass Increase(%)
2,2
2
1,8
1,6
1,4
1,2
1
0,8
0,6
0,4
0,2
0
PANI
PSS
0 5 10 15 20 25 30 35
Number of layers
Figure A-2.1. Mass increase (%) indicating cumulative deposited mass for both cases of
solutions containing salt and without salt as a function of number of layers for sample(B)
The oscillated behavior in Figure A-2.2 is neglected and the linear increase of PANI layers is
taken into account. In 32 deposited PSS/PANI multilayers, the value of the average increase in
thickness is 0.04% for PANI. Hence, for the 19 PANI layers, the mass increase becomes 0.76%.
This means that for a 1g porous sample (HDPE+PANI+PSS), 0.0076g of PANI is available. By
supposing that the density of PANI is equal to 1, the total volume of the blend equivalent to 1g of
HDPE is calculated as:
m
=
ρ
Equation A-2.1 HDPE ( ) 3
V
HDPE
HDPE
1−
0.
0076
= = 1.
167cm
0.
85
Equation A-2.2 VHDPE
0.
333
3
=
⇒ VTotal
= 3.
5cm
V 1
Total
247

Now, as the conductivity of HDPE, air, and PSS polyelectrolytes are very low, it is assumed that
the entire sample, except the PANI network, is void. This implies that the HDPE substrate and
PSS network are considered as porous parts in order to propose a model to predict the
conductivity of the PANI network. Thus, porosity of the sample, including HDPE, voids, and the
PSS network can be calculated as follows:
Equation A-2.3 Vvoid
( 3.
5 − 0.
0076)
Θ =
V
total
=
3.
5
Therefore, a total porosity of 99.78% for this sample is observed.
=
0.
9978
The oldest and most well-known theory of low porosity levels was devised by Maxwell and
predicts the conductivity of CPPD as follows:
Equation A-2.4
⎛ 2 − 2Θ
⎞
σ = σ 0⎜
⎟ = 0.
00147σ
0
⎝ 2 + Θ ⎠
where σ is the conductivity of the porous material, σ0 is the conductivity of the pure material, and
Θ is value of porosity. The porosity of a porous system ranges from 0 to 1. Fricke (Fricke, 1924)
in 1924 added a shape factor due to the geometry of pores in the expression:
Equation A-2.5
⎛ 1−
Θ ⎞ ⎛ 0.
0022 ⎞
σ = σ 0⎜
⎟ = ⎜
⎟σ
0
⎝1
− ( 1−
f ) Θ ⎠ ⎝ 0.
0022+
0.
9978f
⎠
f varies from 1.5 for spherical pores to infinity for thin, disc-shaped pores. Koh (Koh & Fortini,
1973) developed the following equation for highly porous materials with porosity ranging from
0.38 to 0.9:
Equation A-2.6 ⎛ 1 − Θ ⎞
σ = σ 0 ⎜
= 0.
0002 σ
2 ⎟
0
⎝ 1 + 10 Θ ⎠
Skorokhod (Skorokhov, 1972) found that for porous bodies with wide porosity, the equation
would become:
248

Equation A-2.7 0(
) 0
1.
5
σ = σ 1− Θ = 0.
0001σ
Hence, these semi-empirical relations predict the conductivity of porous sample A as low as
0.00147 to 0.0001 times that of pure PANI. Since in our case there is a PANI network with ultra-
high porosity (> 99%), an equation based on measured conductivity data should be developed.
Knowing that the conductivity of pure PANI is around 10 S.cm, and assuming that samples A
and B have similar mass deposition behaviours, from the measured conductivity data the
following equation for the deposition of 38 layers of PSS/PANI (after reaching the stable
plateau) can be given:
Equation A-2.8
σ =
0. 000002 σ
The coefficient obtained from the plot is 50 times less than that of the equation suggested by
Skorokhod (Skorokhov, 1972), due to the disconnection of some clusters in the PANI network.
Skorokhod (Skorokhov, 1972) corrected equation A-2.9 by introducing a relative linear
dimension of contact (ξ) to account for the effect of the imperfection of particle contacts:
Equation A-2.9 ( ) ξ 1.
5
σ = σ −
0 1 Θ
A value of 1.43 was calculated for ξ of sample B through the comparison of equations A-2.8 and
A-2.9. Consequently, the specific relation for the calculation of the conductivity of 3D CPPD on
substrate B can be presented as:
Equation A-2.10 ( ) 14 . 2
σ
=σ −
0 1 Θ
0
249

APPENDIX III
A- 3 Mechanism of Aggregation of Weak Polyelectrolytes
Weak polyelectrolytes have some advantages compared to strong ones, such as the possibility of
tuning their properties by simple pH adjustment. In strong polyelectrolytes, the charges remain
over the entire pH range. On the contrary, for weak polyelectrolytes, the degree of ionization
depends on the pH of the solution. In this thesis, the pH of polyaniline solution is set constant at
2.5 for all samples in order to have the equal fraction of charges for layers. Moreover, the weak
PANI shows a “poor” nature in aqueous solutions(Cheung, et al., 1997). Such poor solvent
quality indicates that the polyaniline solution can only be used for about 1-2 weeks before
important aggregation of the polyaniline chains happens(Cheung, et al., 1997). In this case, a
large amount of polymer chains is adsorbed per layer and multilayer deposition cannot be
controlled. The aggregation of polyelectrolytes in a solution throughout layer-by-layer
experiments was studied over time for sample B in Chapter 5 in two steps (Figure A-3.1). In the
first continuous process, 22 layers were deposited onto the 3D substrate. After the deposition of
the 22 nd layer, all the solutions were left for three weeks to allow the chains to make aggregates.
The second LbL process was continued by the same solutions and substrates, whereas layers 23
to 37 were deposited on the previously prepared multilayer. In this protocol, two different growth
rates of deposition of NaPSS for either step were obtained. The deposition process with fresh
solutions exhibited a growth rate of 3.04, while a higher growth rate value of 6.27 was obtained
for the remaining sample. Such a high growth rate demonstrates the aggregation deposition of
chains in the second step instead of the deposition of a few chains in the first step.
Correspondingly, a solution containing fresh PANI shows a rate of desorption of 2.28, compared
to the solution used in the second step with a rate of 5.22. The aggregation of PANI resulting
from the poor nature of the PANI solution is proposed as the main reason for the increase of the
growth and desorption rates. Therefore, in the second step, a large number of aggregated PANI
chains was deposited, yielding a significant increase in electrostatic force. In the subsequent
250

Cumulative Mass × 10 5 (g)
dipping PSS process, a proportional extent of PSS was deposited to compensate for this
electrostatic force.
180
160
140
120
100
80
60
40
20
0
PANI
PSS
S = 3.04
S = 2.28
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Number of Layers
S = 6.27
S = 5.22
Figure A-3.1. Cumulative deposited mass in LbL process on sample(B) for both PANI and
NaPSS polyelectrolytes as a function of number of layers.
251

APPENDIX IV
A- 4 Relationship between Conductivity and Morphology
of Ternary Polymer Blends
This annex is a comprehensive study of a ternary blend for the case of complete wetting in which
conductivity data are used to justify the behavior of a ternary blend, which is attributed to its
morphology and is based on the composition of its components. It was observed that the lowest
percolation threshold (as low as 3%) for a ternary blend has been found in a double percolated
structure for the complete wetting case. Due to high interfacial tension between PANI and
common polymers, it is almost impossible to situate PANI at the interface.
According to the positive spreading coefficient of PVDF over PANI (λPVDF/PANI = 7.3 mN/m in
HDPE/PVDF/PANI) (Figure A-4.1), PVDF is forced to the interface. In this case, PANI and
HDPE situate in the inside and outside, and PVDF locates at the interface. The lower interfacial
tension between PVDF and PANI results in the interpenetration of the entire PANI and PVDF in
each other, and the formation of small phase sizes. Some big parts of PVDF polymer have been
clearly removed after extraction of PVDF by DMF in such a morphology (Figure A-4.1a).
Solvent extraction/gravimetric measurement reveals that some part of the PANI entrapped in the
PVDF phase is removed with PVDF. By decreasing the amount of PVDF to as little as 20%, as
shown in Figure A-4.1b, the number of large PVDF islands reduces. These disappearances of
large PVDF parts continue with the reduction of the amount of PVDF until no large PVDF parts
are observed in the 45/10/45 HDPE/PVDF/PANI blend (Figure A-4.1c). However, the large
parts of PVDF play an important role in reducing the percolation threshold of PANI, as they
occupy some spaces in the blend and limit the presence of PANI in those areas. It can be
concluded that one of the best ways of reducing the percolation threshold, despite the doublepercolated
structure, is occupying places with other components. In a 33/33/33
HDPE/PVDF/PANI blend, the area occupied by PANI has been limited because of continuous
phases of PVDF and HDPE, and large PVDF areas. This results in a lower percolation threshold
252

of PANI, and an increase in the conductivity of the sample, as for this blend conductivity reaches
a relatively high value of 0.017 S/cm.
a) b) c)
Figure A-4.1. Scanning electron micrographs of HDPE/PVDF/PANI after extraction of PVDF by
DMF (a) 33/33/33, (b) 40/20/40, and (c) 45/10/45
The approach obtained for HDPE/PVDF/PANI can be used to decrease the percolation
threshold of PANI in a HDPE/PEMA/PANI blend. In such a conversion, HDPE plays a major
role as it forms a continuous structure containing large rods and branches, and consequently
limiting the space, whereas PANI can generate clusters only in limited routes. In this part, two
ternary blends of HDPE/PEMA/PANI and PS/PEMA/PANI are studied from electrical and
morphological points of view. It is shown that the high surface tension of HDPE and PANI
results in a positive spreading coefficient: λPEMA/PANI = 8 mN/m in HDPE/PEMA/PANI,
according to the Harkins equation (interfacial tensions are listed in Table A-4.1). Therefore,
PEMA locates at the interface of HDPE and PANI, separating them. In such a morphology, due
to high interfacial tension between HDPE and PEMA, large continuous parts of HDPE exist in
the blend, which restricts the area for diffusion of PANI. Therefore, connected pathways of
PANI are achieved at a lower amount of PANI, leading to a decrease in the percolation threshold
of PANI, and subsequently increasing the value of conductivity. In the ternary PS/PEMA/PANI,
due to a positive spreading coefficient of λPEMA/PANI = 3.2 mN/m, the PEMA phase is located at
the interface of the PS and PANI phases (Table A-4.1). Contrary to the previous case, with a low
amount of PANI, the percolation threshold is not reached due to the low interfacial tension of PS
and PEMA, resulting in PANI droplets remaining in the blend.
253

Table A-4.1. Interfacial Tensions for Polymer Pairs
Theoretical data(Harmonic mean equation) γ (mN/m)
HDPE/PEMA 8.3 PEMA/PANI 10.6
HDPE/PANI 26.9 PS/PANI 15.1
PEMA/PANI 10.6 PS/PEMA 1.3
Ternary polymer blends comprised of HDPE/PEMA/PANI with various compositions of PANI
are prepared. In such samples, the ratios of HDPE and PEMA concentrations are changed while
constant volume fractions of PANI (10%, 20%, and 30%) were selected. Figure A-4.2 shows the
results of resistance for various compositions of HDPE/PEMA/PANI. It is observed that the
binary blend of HDPE and PANI has the highest value of resistance, which is attributed to
droplet-matrix morphology, because of the high interfacial tension between HDPE and PANI. A
percolation threshold of 25% PANI can be detected for this binary blend, as at 25% of PANI, the
value of resistance starts to decrease. It is found that addition of PEMA to the HDPE/PANI
binary blend decreases the resistance of the blend. It is proposed that the interfacial tension
between PEMA and PANI is low, leading to a reduction in the size of the PANI phases in the
blend. Connection of such small phases of PANI and the presence of large branches of
continuous HDPE in the blend result in the formation of small connected pathways, and
subsequently helps to reduce the total percolation threshold. Consequently, addition of PEMA to
the binary blend reduces both the percolation threshold of PANI and the resistance of the blend
for samples with PANI compositions more than 20%. For concentrations of less than 10% PANI,
due to lack of sufficient PANI to form tclusters, other components and morphology play a major
role in determining the value of resistance.
254

Resistance(ohm)
1.0E+12
1.0E+11
1.0E+10
1.0E+09
1.0E+08
1.0E+07
1.0E+06
1.0E+05
1.0E+04
1.0E+03
1.0E+02
0 5 10 15 20 25 30 35 40 45
Volume fraction of PANI (%)
HDPE/PANI
PEMA/PANI
255
PANI/(HDPE8/PEMA1)
PANI/(HDPE6/PEMA3)
PANI/(HDPE5/PEMA5)
PANI/(HDPE3/PEMA6)
PANI/(HDPE1/PEMA8)
Figure A-4.2. Resistance values for ternary blends of HDPE/PEMA/PANI with various
compositions as a function of volume fraction of PANI
Figure A-4.3 represents the same results as Figure A-4.2 from another point of view, since the
resistance has been plotted against volume fraction of one of the components (PEMA). As
expected, Figure A-4.3 shows higher values of resistance for samples comprising 10% of PANI
compared with samples containing 20% or 30% of PANI. Resistance values of samples
containing 10% PANI are represented in the first curve in Figure A-4.3, revealing that increasing
the concentration of PEMA decreases the resistance value of the blend, up to 45% of PEMA.
Further increase of PEMA leads to an increase in the resistance value. This experiment clearly
illustrates the dependence of the conductivity percolation threshold to sample morphology. The
conductivity value decreases when morphology of matrix-droplet for 90/10 HDPE/PANI
changes to matrix/core-shell morphology for the 80/10/10 HDPE/PEMA/PANI ternary blend. At
a composition of 60/30/10 HDPE/PEMA/PANI, shells of PEMA coalesce with each other and
make several connected pathways. The relatively high interfacial tension between HDPE and
PEMA (γPEMA/HDPE = 8.3 mN/m) results in a lower interfacial area value, demonstrating the

continuous structure of PEMA and HDPE with larger phase sizes. Therefore, a large part of the
blend is occupied by thick rods of HDPE. Since the PEMA phase, located at the interface, makes
some connected pathways, and PANI is distributed in the PEMA, in this case geometrical
restriction results in producing a few connected pathways of PANI in 30% PEMA. Because of
the constitution of small clusters of PANI in the PEMA intermediate phase, the resistance
slightly decreases. Increasing further the amount of PEMA results in a tri-continuous
morphology for 45/45/10 HDPE/PEMA/PANI, since with a sufficient content of PEMA, a
continuous phase of PEMA forms and consequently the small amount of PANI phase inside it
becomes continuous. As the continuous HDPE phase still occupies 45% of the blend, and due to
the formation of larger connected pathways of PANI, resistance reduces by almost three orders
of magnitude. Addition of further PEMA to the blend results in a preparation of 30/60/10 in
which the continuous structure of HDPE starts to convert to droplets. Due to the lower
occupation of space with HDPE, and the dispersion of PANI in a higher amount of PEMA, 10%
PANI distributes as droplets and connected pathways are hardly constructed. Increasing the
concentration of PEMA to 10/80/10 HDPE/PEMA/PANI results in the formation of
bicontinuous/droplet morphology, whereas a small content of HDPE causes a vast occupation
area of PEMA. Therefore, 10% PANI particles disperse in PEMA, showing a higher value of
resistance. Finally, for a binary blend of 90/10 PEMA/PANI, matrix/droplet morphology is
predicted, although still one or two connected pathway of PANI in PEMA can be found. As it is
observed, concentration and morphology of HDPE and PEMA affect the conductivity and
continuity of PANI, where the percolation threshold of samples with a constant 10% of PANI is
obtained at 35% of PEMA and 90% of PEMA in a HDPE/PEMA/PANI ternary blend.
In samples containing 20% of PANI, a similar behavior for small contents of PEMA with a slight
difference in the value of resistance is observed. At 40/40/20 HDPE/PEMA/PANI, tri-continuous
morphology is generated, demonstrating a sharp decrease of the resistance. The important points
are samples containing 20% PANI and 66.6/33.3 PEMA/HDPE and 90/10 PEMA/HDPE in
which, in contrast to samples containing 10% of PANI at these points, resistance has a low value.
As mentioned above, higher value of resistance at these points for samples containing 10% of
PANI is related to droplets of PANI distributed in PEMA, due to lack an insufficient amount of
PANI. On the contrary, for samples containing 20% of PANI, the concentration of PANI is high
enough to form a number of connected pathways of PANI inside PEMA, yielding the decrease of
256

esistance value. Samples containing 30% of PANI, because of the high amount of PANI in the
sample, reveal high values of conductivity for a wide range of concentrations. Only in the first
region of the curve (high concentration of HDPE), due to the presence of a large volume fraction
of HDPE, is the value of conductivity low. The percolation threshold occurs at 10% of PEMA, as
a double-percolated morphology is predicted for this ternary blend containing of 60/10/30
HDPE/PEMA/PANI. By increasing the amount of PEMA, the morphology tends to become tri-
continuous, since the network of PANI contributes throughout the entire sample. The second
ternary blend, comprised of PS/PEMA/PANI, was chosen with the same compositions of
components. Since the interfacial tension between PS and PEMA (γPEMA/PS = 1.3 mN/m) is much
lower than that for HDPE and PEMA (γPEMA/HDPE = 8.3 mN/m), it is predicted that PS and PEMA
will make smaller phase sizes than HDPE and PEMA. Although it is difficult to show the
morphological features of a ternary PS/PEMA/PANI blend, the decrease of the phase sizes
between PS and PEMA are shown in the HDPE/PS/PANI blend. Figure A-4.4 shows the
morphology of an interdiffusion of PEMA and PS in the HDPE/PS/PANI blend, representing
very small rods and branches of PEMA after extraction of PS, while the rods of the continuous
phase of HDPE are so large. Such differences in the sizes of trunks of continuous HDPE and
PEMA indicate how much the interfacial tension differs between PEMA and either of these two
phases.
257

Resistance (Ohm)
1.0E+12
1.0E+11
1.0E+10
1.0E+09
1.0E+08
1.0E+07
1.0E+06
1.0E+05
1.0E+04
1.0E+03
1.0E+02
PANI = 10% of total PANI = 20% of total PANI = 30% of total
0 10 20 30 40 50 60 70 80 90 100
Volume Fraction of PEMA (%)(HDPE+PEMA=100%)
Figure A-4.3. Resistance values for ternary blends of HDPE/PEMA/PANI comprising of 10%,
20%, and 30% of PANI with various compositions ratio of PEMA and HDPE as a function of
volume fraction of PEMA
Figure A-4.4. Scanning electron micrographs of 40/10/50 HDPE/PS/PEMA compounded by
melt-blending after extraction of PS by cyclohexane
258

The results of resistance for PS/PEMA/PANI are plotted as a function of volume fraction of
PEMA in Figure A-4.5. The plot of resistance values for PS/PEMA/PANI samples containing
10% of PANI is completely different from what was observed for HDPE/PEMA/PANI,
represented in Figure A-4.3. Figure A-4.5 reveals that for blends containing 10% of PANI at all
concentrations of PEMA, the resistance value is significantly high, since for all concentrations of
PEMA the resistance is approximately constant and the percolation threshold is not reached. The
reason for not achieving the formation of PANI connected pathways in blends with a low amount
of PANI (10%) is proposed to be attributed to the low value of interfacial tension between
PEMA and PS. Such a low interfacial tension results in the constitution of a co-continuous
network of PEMA and PS, leading to dispersion of droplets of PANI in a large volume. In the
previous case (HDPE/PEMA/PANI), because of many parts of blend being occupied by HDPE,
PANI was forced in geometrically restricted paths, and a few connected pathways of PANI, even
at low concentration, are formed. In the PS/PEMA/PANI case, there is no force to get the PANI
droplets coalesced because of the interdiffused network of PEMA and PS resulting in the
distribution and dispersion of PANI in the droplet shape throughout the sample. Such
morphology represents unconnected droplets of PANI, and subsequently, due to the absence of
formation of a PANI connected network, a high value of resistance for all compositions of
PEMA is revealed. However, for samples containing 20% PANI, due to the sufficient amount of
PANI at a high content of PEMA, networks of PANI are formed, and for this plot a percolation
threshold of 20% of PEMA can be considered. High concentrations of PS reveal a matrix-droplet
morphology, as the value of resistance is high.
259

Resistance (Ohm)
1.0E+11
1.0E+10
1.0E+09
1.0E+08
1.0E+07
1.0E+06
1.0E+05
1.0E+04
1.0E+03
1.0E+02
PANI = 10% of total PANI = 20% of total
0 10 20 30 40 50 60 70 80 90 100
Volume fraction of PEMA (%)(PS+PEMA=100%)
Figure A-4.5. Resistance values for ternary blends of PS/PEMA/PANI comprising of 10% and
20% of PANI with various compositions ratio of PEMA and PS as a function of volume fraction
of PEMA
It is concluded that the composition of the inner and outer phases in a double percolated structure
can either be effective or ineffective, depending on the interfacial tension of the phases.
References of Annexes
Berger, W., Kammer, H. W., & Kummerlowe, C. (1984). Makromol Chem Suppl B, 8, 101.
Caruso, F. (2003). Hollow Inorganic Capsules via Colloid-Templated Layer-by-Layer
Electrostatic Assembly (pp. 145-168).
Caruso, F., Caruso, R. A., & Mohwald, H. (1998). Nanoengineering of inorganic and hybrid
hollow spheres by colloidal templating. Science, 282(Copyright 1999, IEE), 1111-1114.
Caruso, F., Caruso, R. A., & Mohwald, H. (1999). Production of Hollow Microspheres from
Nanostructured Composite Particles. Chemistry of Materials, 11(11), 3309-3314.
260

Caruso, F., & Schuler, C. (2000). Enzyme Multilayers on Colloid Particles: Assembly, Stability,
and Enzymatic Activity.. Langmuir, 16(24), 9595-9603.
Cheung, J. H., Stockton, W. B., & Rubner, M. F. (1997). Molecular-level processing of
conjugated polymers. 3. Layer-by-layer manipulation of polyaniline via electrostatic
interactions. Macromolecules, 30(9), 2712-2716.
Fricke, H. A. (1924). mathematical treatment of the electrical conductivity and capacity of
disperse systems. I. The electrical conductivity of a suspension of homogeneous
spheroids. J. Physical Rev.,, 24, 575-587.
Gubbels, F., Jerome, R., Teyssie, P., Vanlathem, E., Deltour, R., Calderone, A., et al. (1994).
Selective localization of carbon black in immiscible polymer blends: a useful tool to
design electrical conductive composites. Macromolecules, 27(7), 1972-1974.
Hobbs, S. Y., Dekkers, M. E. J., & Watkins, V. H. (1988). Effect of interfacial forces on polymer
blend morphologies. Polymer, 29(9), 1598-1602.
Hou, S., Harrell, C. C., Trofin, L., Kohli, P., & Martin, C. R. (2004). Layer-by-Layer Nanotube
Template Synthesis. Journal of the American Chemical Society, 126(18), 5674-5675.
Koh, J. C. Y., & Fortini, A. (1973). Prediction of thermal conductivity and electrical resistivity
of porous metallic materials. Int. J. Heat Mass Transfer,, 16, 2013-2021.
Lyngaae-Jorgensen, J., & Utracki, L. A. (2003). Structuring polymer blends with bicontinuous
phase morphology. II. Tailoring blends with ultralow critical volume fraction. Polymer,
44(5), 1661-1669.
Narkis, M., Haba, Y., Segal, E., Zilberman, M., Titelman, G. I., & Siegmann, A. (2000).
Structured electrically conductive polyaniline/polymer blends. Polymers for Advanced
Technologies, 11(8-12), 665-673.
Omonov, T. S., Harrats, C., & Groeninckx, G. (2005). Co-continuous and encapsulated three
phase morphologies in uncompatibilized and reactively compatibilized polyamide
6/polypropylene/polystyrene ternary blends using two reactive precursors. Polymer,
46(26), 12322-12336.
Park, M.-K., Deng, S., & Advincula, R. C. (2005). Sustained Release Control via Photo-Cross-
Linking of Polyelectrolyte Layer-by-Layer Hollow Capsules. Langmuir, 21(12), 5272-
5277.
Potschke, P., & Paul, D. R. (2003). Formation of co-continuous structures in melt-mixed
immiscible polymer blends. Journal of Macromolecular Science - Polymer Reviews,
43(1), 87-141.
Reignier, J., & Favis, B. D. (2000). Control of the subinclusion microstructure in
HDPE/PS/PMMA ternary blends. Macromolecules, 33(19), 6998-7008.
261

Reignier, J., & Favis, B. D. (2003). Core-shell structure and segregation effects in composite
droplet polymer blends. AIChE Journal, 49(4), 1014-1023.
Reignier, J., Favis, B. D., & Heuzey, M.-C. (2003). Factors influencing encapsulation behavior
in composite droplet-type polymer blends. Polymer, 44(1), 49-59.
Schneider, G., & Decher, G. (2004). From Functional Core/Shell Nanoparticles Prepared via
Layer-by-Layer Deposition to Empty Nanospheres. Nano Letters, 4(10), 1833-1839.
Skorokhov, V. V. (1972). The Rheological Principles of Sintering Theory. Kiev: Nauka.
Soares, B. G., Gubbels, F., Jerome, R., Teyssie, P., Vanlathem, E., & Deltour, R. (1995).
Electrical conductivity in carbon black-loaded polystyrene-polyisoprene blends. Selective
localization of carbon black at the interface. Polymer Bulletin (Berlin), 35(1-2), 223.
Steinmann, S., Gronski, W., & Friedrich, C. (2001). Cocontinuous polymer blends: Influence of
viscosity and elasticity ratios of the constituent polymers on phase inversion. Polymer,
42(15), 6619-6629.
Torza, S., & Mason, G. (1970). Three- phase interactions in shear and electrical fields. 33(1), 67-
83.
Valera, T. S., Morita, A. T., & Demarquette, N. R. (2006). Study of morphologies of
PMMA/PP/PS ternary blends. Macromolecules, 39(7), 2663-2675.
Van Oene, H. J. (1972). Modes of dispersion of viscoelastic fluids in flow. Colloid Interface Sci.,
40, 448-467.
Virgilio, N., Favis, B. D., Pepin, M.-F., Desjardins, P., & L'Esperance, G. (2005). High contrast
imaging of interphases in ternary polymer blends using focused ion beam preparation and
atomic force microscopy. Macromolecules, 38(6), 2368-2375.
Volodkin, D. V., Petrov, A. I., Prevot, M., & Sukhorukov, G. B. (2004). Matrix Polyelectrolyte
Microcapsules: New System for Macromolecule Encapsulation. Langmuir, 20(8), 3398-
3406.
Yuan, Z., & Favis, B. D. (2004). Macroporous poly(L-lactide) of controlled pore size derived
from the annealing of co-continuous polystyrene/poly(L-lactide) blends. Biomaterials,
25(11), 2161-2170.
Yuan, Z., & Favis, B. D. (2006). Influence of the efficacy of interfacial modification on the
coarsening of cocontinuous PS/HDPE blends during quiescent annealing. Journal of
Polymer Science, Part B: Polymer Physics, 44(4), 711-721.
Zilberman, M., Siegmann, A., & Narkis, M. (2000). Melt-processed, electrically conductive
ternary polymer blends containing polyaniline. Journal of Macromolecular Science -
Physics, B39(3), 333-347.
262